Nope, it explicitly includes the art score twice.
The puissant alters the effective art score. It's explicit.
Because you aimed it, and where you hit them affects the damage.
With magic, your accuracy doesn't affect the damage. It just doesn't in the ars magica system. That's your difference.
I'm not making them make aging rolls DURING an adventure, so that scenario isn't going to happen.
Again: This is one of the situations where it's actually a hard thing to interpret.
Hmmm, assuming only 4 botch die (which is completely implausible in the circumstances, I'm thinking 11 botch dice would be more appropriate) you're probably (more than 93% chance) botching; due to the "roll twice" result being a 90% chance each time. With a more realistic number of botch dice, you're looking at a 96% chance.
Oh, and you're probably getting a lot* of botches. You have a (far) better chance of reaching final twilight (starting from 0 warping points) and/or dying than of actually succeeding at your experiment.
That doesn't seem like a great trade-off to me.
(*by which I mean "More than the number of atoms in the universe". Your average number of botches is infinite.)
Of course, personally I'm of the opinion that at -7 the "no risk of worsening" is irrelevant: you literally can't do the labwork. It's not that it'd strain you too much, it's that it's literally impossible for you to do with that huge a penalty.
Well, divine magic resistance is neither a roll nor a total, so I'd have to say no.