No, I wouldn't say that's very accurate. In part because it treats IQ as a single, monolithic number. A child can have PG scores in one area and have the overall IQ score pulled down by things less tied to GTness (like Processing Speed, since smart does not always equal fast), yet still be bored silly in class. (BTW, that's where the GAI comes in. It leaves out those scores that aren't as closely tied to GTness.)
My DS6 has an overall FSIQ score just under 140, but with a PRI that was well into in the upper range that you listed. He was bored, angry, miserable and even began acting out in first grade, which was totally unlike him. All of this is why we had his IQ tested. When we saw his scores and looked at the behavior we had been seeing, everything made sense. We pulled him out of public school for homeschooling. If he were to go back to our public school, he would almost certainly have to be grade skipped. But our choice had MUCH more to do with his level of happiness than with his scores. The scores just explained the unhappiness.
Then there are kids who are HG+, with scores in that upper range, who are very happy in a regular age-based classroom. Granted they do usually require significant adaptation--at least subject acceleration--but if the regular classroom meets their needs, those kids are very happy.
Scores are one part of the puzzle that can help to make sense of what a kid needs and why. But making decisions based solely or even mainly on one score on one test on one day isn't very sensible. If you're just looking for trends, maybe you could use that as an argument with the school, but it's not something I'd want to stand behind personally. FWIW...
As for GAI over FSIQ--what's the spread between the subtests? GAI is very useful if there's a large spread somewhere.
I don't know if this helps...
