I'd also point out that the criticisms leading to new models or new experiments are *crucial* for the development of science. Sometimes it *does* happen that the new experiments don't turn out as everyone expects. Sometimes the new models *don't* give the answer everyone has thought they would.
A very good example was the Michelson-Morley experiment testing the theory of the ether. At the time, this explanation for the propagation of light was considered incredibly solid and *nobody* expected the result of the experiment to say anything different than this theory predicted.
Well, of course, the result *didn't* go as expected. That ultimately caused a radical re-thinking of our ideas of how light propagates and the development of special relativity by Einstein.
So, one of the *jobs* of a research scientist is to point out leaps of logic others have committed (or that they themselves have) and find ways to *test* to see whether those leaps actually are over holes. In the vast majority of cases, they are not, of course. But every once in a while, something turns out that requires us to do more work.
Such times actually will make the career of the person who pointed out the flaw and the person who verified it. This is how scientific revolutions are born.
But what never happens is that we go back to the *old* explanation, the one prior to the current one, because that one was rejected because it failed to match observations. This is why the basics of evolution and common descent aren't going to go away. There is simply too much that has been verified.
I might point out that, even after the Michelson-Morley experiment, the *equations* that describe the propagation of light remained the same. What changed was our understanding of what those equations said and meant.