I will say at the bare minimum if you aren't going to use AAMC as full length tests you really need to have at least several practice tests that are full length and simulated under real conditions with other companies. You can't say you prepared to the best of your abilities otherwise.
But to go off what was said before what's important is to really go through every little piece of analysis the AAMC gives of their tests. For answers that are clearly wrong and answers that are clearly right. There is a reason why they choose to put them there and if you analyze every single answer choice, why it is wrong, why it exists(ie what it says and what kind of trick is in it amongst other things), you will pick up on trends and be better for it. This is a test as much of anything to think like the test makers themselves. Go through every single choice of as many AAMC practice tests you can get your hands on, old MCAT and new MCAT.
The verbal is very very similar new test to old test; if you analyze through the 10-11 available practice tests old and new from AAMC each of the verbal passages and each answer choice(why they are wrong, what type of answer they are, the type of trick the AAMC is trying to pull with etc) that's how you really start to think like the test and pick up on what they do. You will have over 400 verbal questions from passages to analyze for this. That is a pretty decent sample size to start picking up on things the AAMC likes to do and see through their logic. Of course this was better done back when the old AAMC tests were online with answer keys posted by the AAMC to directly see their logic, but alas.
This is the type of thing that will involve going through questions and answer choices repeatedly. Forget about whether or not you remember the correct answer when you go through it again. It's largely irrelevant for this. The key is to see the logic behind each and every answer choice and start to pick up on patterns. This can only be done through looking through these tests a number of times thoroughly.