The second quoted paragraph seems to be attacking a strawman. I don't think it was suggested that we should add silly details to improve falsifiability, but rather remove them. Moreover, it seems like this is a way to choose between existing theories, rather than a way to mutate one theory into a better one.
It's pointing out that the equivalence of "simpler" with "more falsifiable" is not perfect. Nobody is suggesting that we just add silly details for the sake of increasing falsifiability, but suppose two research groups independently arrived at those competing theories. Should we choose the simpler one or the more falsifiable one?
