Eh, check the SAT scores. The difference is 0.14 points favoring men on the 1600, 29 points (about a tenth of a standard deviation) favoring women on the 2400, and .44 points (about a tenth of a standard deviation) favoring women on the ACT, and 3.7 points (about a quarter of a standard deviation) on the IQ self-report. I wouldn't trust the IQ numbers enough to rest an argument on that point.

I would also state the argument a bit differently--it's not that at equal levels of IQ more women are filtered out than men (in which case the IQ distributions of the two would only be different matching any underlying population differences), but that the IQ filtering effect is stronger for women than men (or, stated symmetrically, less women are filtered out at higher IQs than lower IQs).