It would be interesting to know, @InfinityzeN, what balance you have had between uptime and downtime in the sagas sampled.
There is no doubt that if you design a magus for stage light, Intelligence is rarely going to be the most valuable char. Most players I have seen, however, make their characters for the long game with downtime dominating, and there is not so much you can do but lab work.
Most of them have been slanted towards down time, though the speed has ranged from slow to fast. You are correct in that both Int 0 characters have happened in very active games with little downtime, though there have been many with Int 1 or 2 in the more downtime slanted games.
In the long run, Int makes up such a small chunk of Lab Total (with excess often wasted) that there is little effective difference between a 1 or 2 verse a 3+ outside of Magi who do things like invent high level spells or constantly try to enchant items at the limits of their capabilities (or do things like Original Research). A Magus who primarily reads books and learns spells from text will most likely not really notice the difference.
(EDIT: I am talking about my personal experience in Sagas I have played in. How valuable Int is to your Lab Total can vary drastically based on the Saga, especially in ones which are book/spell text poor and make getting a high lab total difficult. So very much YSMV.)
Maybe it is just me who prefer to operate at the limits of capabilities.
You are not alone there. Bonisagus is my fav House and I tend to like playing a Magus who pushes the limits as much as possible. However all the groups I have played AM with have included a fair number of players who enjoyed characters designed for things other than "Optimal Lab Research".
In previous editions (3rd and I think 4th) Int affected how much xp one gained from seasonal activities so it was much more important and I almost never made average or worse Int magi, usually veering towards +3.
In 5th I probably average giving my magi +2 with a significant number being +1 or +3 and I almost never increase any stat to +4 or +5 in character generation as the opportunity cost seems huge vs any other minor virtue and may be improved later and usually much more easily than initiating many virtues. But while the bonus to lab total is relatively small it, like MT, affects every lab activity so it is actually more powerful than you seem to be suggesting.
That was 4ed. Insanely complex and unbalanced rules. No wonder they only lasted one edition 
3ed only had story xp, plain and simple
I was scratching my head over that one since I played a lot of 3rd and pulled out my book to look. In 3rd story xp, learning Arts or Magic Theory from books or vis is completely unaffected by Int. Learning spells is since it is based on Lab Total, but it is a small portion of the total that gets smaller as the Magi become more powerful.
4th I have the least experience of the editions I have played with, though it does have a few wonky XP rules that advantage Int. Being Trained in a Skill gains +1 XP if your Int is positive (no difference between 1 and 5), being Trained in Spells gets your full Int added for calculating if you can learn it, Practicing Spells gains Int in XP, Exposure gains 1 XP for Int 0 or less and 2 XP for Int 1+, Disputatio gains Int from both teacher and taught, Reading books added both Int and Concentration to the study total, etc.
That my experience with games is mostly 3rd and 5th, with a few 2nd and only one 4th, that would explain my view that high Int is not extremely valuable. 4th seemed to add a whole lot of "High Int is IMPORTANT!" rules for XP, along with much more complexity than other editions.
EDIT: Yes I know and understand the value of Int to Lab Total in 5th. As stated earlier, Bonisagus is my fav House. However many of the players in the Sagas I have been part of did not prioritize Lab Work or a high Lab Total. Even in my current 104+ year Saga there are several characters who spend their down time mostly reading/writing books, with only the odd season learning a spell or enchanting an object where having a Lab Total that is 1 or 2 points less did not negatively affect them.
How valuable that 1 or 2 points of Lab Total to you is very much YSMV and YCCMV (Your Character Concept May Vary). If you can increase Int during play by Ritual or Virtue it becomes even less important at the start of play.
No impact to Lab Total, but "no magus can use more than twice his Magic Theory score in pawns of vis in a season". That's where low Int becomes painful.
How? Your Magic Theory score does not depend on your Int.
Brain fart?
To me there are a few obvious "uses" for low int gifted magi:
The other side of this is the cat/string equation (the reason that cats are smart is the dumb ones died out with the invention of string), between being social outcasts and having a poor ability to harness whatever supernatural ability they attract the survival of low intelligence Gifted is likely to be rather low...
Honestly in my experience people are more fixated on a high stamina than a high int.
I didn't realize that people saw +3 INT as practically required.
Eh, +3, not hardly. Positive, yes, not quite absolutely but darn close, in my (admittedly limited) experience.
Prioritizing stamina makes a lot of sense, though.
The "Usual Intelligence score for Magi ... is +3" is a quote taken from HoH:S rather than something that seems to hold true in most of the games commented about in this thread.
The template Magi in the Core ArM5 book average just over +3 Int, with 25% having Great Characteristic (only two in Int) and 33% having Improved Characteristics.
The template Magi in 4th do not include Characteristics. However it has the wonky XP rules which make Int a priority for many things, which most likely resulted in a spike of high Int Magi. It is the only edition with this type of XP setup.
The template Magi in 3rd edition average under +2.5 Int (if you ignore Mercere at Int -1 who is an Ungifted Redcap).
2nd edition does not have template Magi or even Houses in the core book. The Houses were added by the "Order of Hermes" supplement.
of course logically most mystagogues should be Jerbiton with the gentle gift and high presence... but clearly that isn't how it happens since most mystagogues are in the mystery houses (which might, therefore, frequently favor presence over stamina or intelligence)
This is the kind of cultural beliefs which easily develop in a group, and which can develop very differently in different groups.
That inference immediately sounded bizarre to me, but it took quite some thinking to realise why it would be bizzare.
The connection is probabilistic rather than deterministic. Gifted people with -3 Int exist, but they are rare. CharGen rules which capture that statistical correlation would have two problems. Firstly, they would be over-complex and unplayable. Secondly, players should be encouraged to play the statistical outliers. Therefore, this is left out of the rules, and the troupe can populate the world as they please.
I agree. To elaborate on the point you make:
The rules arent created in such a way that if you used them to create a character sheet for every person in ME then you would get the population average. Thus the rules dont force people with the Gift to have higher intelligence even though it is stated that these two correlate.
The rules exist to make the creation of a character as easy an enjoyable as possible often cutting corners or ignoring things that are "generally true" in the world if favor of achieving the aims of the game.
Thus we cannot assume that the rules always predict information about how the game world functions.
Totally. If the PC generation applied to everyone in ME then the average characteristic would be slightly under +1 and 0 would no longer be average.
If they are, say, Int 0-1, they probably join House Jerbiton, for which you don't need to be clever, you need to be fun to have around and have some skill in arts.
Generally, though? They don't exist.
The point of making the Gift rare, and setting it so that all people with it are desirable to the Order is to lower the implicit infanticide rate caused by the Gift.