I forget the max GPA in our class, it was somewhere between 4 and 5, but I don't think it was the theoretical max you could get if you took as many AP classes as possible and got A's in all of them.
Colleges weren't stupid about it, either, though. My brother had a lower GPA since he went to a smaller school with very few AP classes available, but did better relative to that curriculum than I did relevant to mine's, and he got more scholarship offers and quicker admission offers at the schools we both applied to. Even though my SAT was 10 points higher.
Low IQ would mean that studying wont help you - leading to low results consistently. High IQ means you are capable of good results. The rest is down to a lot of factors that are not IQ.
Since IQ+Conscientiousness is better than either alone, of course they'll want to consider both.
This is problematic if you want it to be a valid measure because it lets people artificially close the gap.
Do you want the student with such a well-rounded education throughout their life that they got a >=99.9% score without needing to study, or the one who caught up by grinding for a few weeks without that same broad underlying fundamental base? If your test can't tell that apart, it's flawed, because access to the right prep is going to become a significant part of what you're measuring, and that access is (for obvious economic reasons) distributed along unequal lines, reinforcing the already-fortunate.
My suspicion, since I doubt we can come up with a test to prep, is that the best answer is either (a) nobody preps or (b) everyone gets access to a lot of prep. The former seems impossible to enforce, plus is the sort of "individual liberty"-constraining thing that the politically dominant swathes of America hate. They want to be able to make their kids the best, they "earned it." The latter would be easier, but isn't without some complications, and of course is time that could be better spent on something other than maximizing a single metric.