Those are some good insights.

Science, in its simplest form, is a way of applying reason and experimentation to find out what's true.

Discard experimentation or the idea of an objective reality and you end up where mathematics and other "fuzzy" fields sometimes end up, in the same place as things like "Zeno's Paradox."

Discard reason and analysis, and you're left with (basically) random trial and error. This can work, and has, but without analysis it's a slow process, and you can't easily come up with new abstract ideas outside of that trial-and-error process. Edison had to try hundreds of filament materials for a light bulb. Today, it might be possible to incorporate analysis of things like molecular structure and physics to predict what materials might make the best structures and go from there.

Mathematics in practical use works because it describes reality accurately. A tool to harness abstraction, as you say. But when reality doesn't match up, the math should change, not the other way around. Einstein's theories were accepted because they matched unusual real-world observations at the limits of things, and still continue to do so as new methods of observation are invented. I worked on Gravity Probe B.

When people start trying to bend reality to match the theories, in ALL fields, things start to go wrong. In "fuzzy" fields of study (I'm open for suggestions for better terms - this one's from engineering schools) it is much easier to do this and get away with it for a long time than those fields where reality comes down on you like a brick (sometimes literally) if you get it wrong. I was in the School of Architecture, which straddles the engineering and "fuzzy" zones, so these things stood out quite a bit.

The idea that "nothing is real" is sometimes a last-ditch human reaction to reality not matching the "math."