The issue is not the average, but the extremes. When you have two identical distribution curves but one's center is shifted, the effect on the ends of the distributions are exaggerated.
A standard deviation increase in the average leads to a huge difference in demographics at the extremes. The population with lower mean will have the majority of the lower tail of the distribution, and the the other population will dominate the upper tail. In the highest 1% they will appear enormously over-represented.
In the middle is where it matters least, because near average people for both populations might have very similar iqs.
For a paper on the issue of specifically ashkenazi (rather than 'jewish') IQ I recommend this one from the University of Utah:
http://web.mit.edu/fustflum/documents/papers/AshkenaziIQ.jbi...
Note the increased IQ is specifically for askenazi jews, as opposed to say sephardic jews. The paper identifies
historical social issues that would make selection for intelligence a strong factor in ashkenazi history, explores possible genetic causes identified through unique mutations not present in adjacent populations, and goes into depth on IQ and IQ heritability.
A high level summary available from the economist behind a paywall but viewable here:
https://outline.com/XnhPWd
I hope you find it interesting. I think its scientific enough that few people should find it controversial, except those who believe that IQ is not a useful measurement of intelligence, which I don't think is an opinion you share given your previous posts.
The controversial part would be Jordan B Peterson talking about it here:
https://www.youtube.com/watch?v=JQHH6o9ual8
No matter your feelings on him, worth a watch.