While "computer science has as much to do with computers as astronomy does with telescopes", I fear that it's become too inexorably bound to computers and too dependent on its foundation in the maths. As a result CS is seen as sort of a fancier software developer/engineering field and not a field that's about theories of computation.
I think CS has the potential to reframe a huge number of fields and offer a different or unique take on how those fields might approach problems. e.g. a chemical reaction could be reframed as a computation that takes inputs and produces outputs. Biology, economics, finance, psychology, etc. could take on entirely new aspects if viewed through a computational lens. I also think that such a lens could enable higher-order multidisciplinary, interdisciplinary, and ultimately convergent efforts [1] by simplifying the reasoning across disciplines -- defeating barriers while allowing disparate experts to work better together. CS has also developed a number of very powerful tools for encoding, aggregating, abstracting, executing, and validating the collective thoughts of massive numbers of individual contributors -- I'm not sure any other discipline is able to organize millions of atomic computational steps, all provided by people who have never met, test them, build them, and execute them.
Imagine needing to develop a complex system, one that might draw from dozens of disciplines and have millions of atomic steps to achieve, and the entire effort is organized under atomic instructions that each drive some step, but they could come from any discipline. We (Computer Scientists) have already figured out how to manage that complexity fairly reliably, version it, change it, etc. "Build a car" could be source code somewhere in a repository, and when you need to make one you "execute it" and suddenly mines in Brazil start extracting ore, fabs in Korea start making touch screens, and in a little while you get a car. Somebody could decide to adjust how the blinker works, push the change, it gets reviewed, unit tests "run" in some blinker factory somewhere, and now the future versions of those cars get the change.
1 - https://research.ncsu.edu/rdo/the-difference-between-multidi...