It is no surprise to any classical economist that actors do not always act rationally and the foundation of microeconomics is not built on the unstable platform of assuming that everyone IS rational; but rather that treating actors AS rational is a just and fair way to act while also removing sources of bias.
It does not mean that rules on how to approach people should NOT be modified via analysis of behavior, but rather that a priori we should assume people act rationally in their own self-interest as a first principle and go from there, not assume that everyone is irrational and create a ten million line case/switch statement.
No, economics is a predictive social science, not a system of moral philosophy; the rational choice model has nothing to do with how it is just and fair to treat people. It was the dominant model because despite being self-evidently not fundamentally correct (it assumes a number of things, like perfect knowledge of future outcomes, that no one believes to be literally true or even plausible) it was believed to be an approximation that would predicted results better than any alternative model available at the time, in general and in a wide array of specific cases.
The entirety of it is not. Not by a long shot.
It's also been my experience that when I hear people talking about one another as irrational, it's never because they're trying to be more respectful, but is invariably in support of patronizing or manipulating people. Nobody has ever used an irrational actor hypothesis to justify paying someone more, or giving them more authority and autonomy.
Amusingly, I've noticed that most HR managers have psychology degrees. I've wondered: What if HR managers had economics degrees. Would we be paid more?
When I was a manager, I received some training from the HR department, and we were constantly reminded that our employees didn't really want to get paid more. This was based on "research." I wondered if this was the result of the researchers simply delivering the result that their patrons wanted to hear.
I've decided to favor an economic view over a psychological one: Be nice to people, of course, but treat them as homo economicus, behaving rationally in regard to their own self interest. While this view isn't proven, I think it's better than making assumptions about their psychological motivations that are likely to be biased by factors such as social class, age, gender, and possibly race and ethnicity.