Thank you short stripper, that was the point I was trying to put across, in normal ops with a 'normal' flat four aircooled engine, I don't see any issues, but in the examples outlined above, I can very much understand how it could be a problem.
What I meant about the difference with large radials is not that there operating temps are different, rather that when the power is reduced, because of the very way they have to able to get vast amounts of air through them to keep their operating temps sensible, that same ability would cause them to cool faster than a well cowled flat engine. My only experience of radial engines is in a Yak 52/18 and even then it is limited, but the cowling fins had a great amount of importance put onto them.
Despite my comments about whether shock cooling is really an issue in normal light a/c, I do operate them as if there was a problem. Since I'm not really convinced one way or the other because of no actual test data, (I'm still sure about the drop off rates in a recently shut-down engine, but I can't find the results of an experiment from university that I was part of. A slightly different thing really, as it was measuring temperature differentials across an water-cooled car engine) I operate with due caution and teach the same.