Maybe it would be better to say the more stunning the breakthrough the more dissimilar the fields.
It's kind of like splicing fruit or cross breeding species. There comes a point when it ain't gonna happen.
The number of advances probably increases as the fields become more similar, at least to a point. The only problem is that the advances aren't as great.
Fast nickel v slow dime optimization problem.
The N. A. of Sciences needs to develop some kind of units of "distance" between two fields, say chemistry to physics is one "ID", to generate all kinds of statistical data, plots of breakthroughs v ID etc.