Thought experiments can reveal profound results beyond what current instruments are capable of measuring.
In 1935, Einstein and friends published a thought experiment that basically said "Well if all this quantum theory is true, then locality can't possibly be a principle." It turns out that locality isn't actually a principle of physics and now we have entanglement, Einstein's "spooky action at a distance."
To the extent that thought experiments suggest real experiments, they are indeed valuable. I disagree that they can 'reveal ... results', however. For instance, Einstein's thought experiment was designed to 'prove' that QM was wrong. Obviously, that result was incorrect.
My other thought is that once you're doing 'thought experiments', you've become rather unmoored from science as I understand it: Observation, Hypothesis, Prediction, Experiment, and so forth. Hopefully, the 'thought experiments' will lead to something testable.