The basic injunction is sound. One must certainly avoid making too much of what might well be superficial similarities. But the methodology necessary for critically examining such similarities to determine whether they are, indeed, superficial -- or not -- has for a great many years been actively discouraged. I'm talking, of course, about comparative musicology (not to mention Cantometrics). Nketia was already going way out on a limb, even in 1962, in claiming that essentially the same procedure could be "commonly found in African musical practice" generally. To go on to even consider the possibility that African and Medieval European hocket could have common roots would have been entirely too much.
As far as Ethnomusicology is concerned, the situation has only worsened over the years, to the point that all but the most narrowly focused comparative research is rarely seen anymore. However, the situation with respect to the rest of the world has changed. Dramatically! Thanks largely to important breakthroughs in anthropological genetics, all sorts of possibilities for considering and reconsidering various similarities and patterns -- of social structure, archaeology, culture, language, etc. -- have arisen and are being enthusiastically explored. Only musicology is lagging behind, saddled with dogmas that have become hopelessly outdated and clearly irrelevant.