This metaphor can be extended the opposite direction. Most people could get much better performance out of their car by training regularly. Isn't a racecar driver a much better driver than your average commuter? Yet you'll notice that while some people take the occasional racing car class for funsies, the actual number of racecar drivers (hobby or professional) is very small.

The issue is opportunity cost. Most people have other things they'd rather be doing, so they learn the minimum they need to accomplish their tasks and get on with it. Fundamentally, most of us who know how computers work deeply didn't learn how to do all this stuff to use computers better, we did it because we enjoyed it. Not everyone enjoys learning how computers work, and that's ok.

Okay, so a car can do two things not one thing: race and commute. A computer is obviously still a much more flexible machine.

Furthermore, as the description of the assessment mentions, "The reasoning strategies that underlie computational thinking can help make sense of complex ideas and solve problems."[0]

Learning about computers isn't just learning an extremely flexible tool; it's learning how to think in a rigorous, abstract manner. It's not the only way to learn critical thinking skills, but it's not possible to learn about computers without them.

[0] https://nces.ed.gov/surveys/icils/about.asp

Cars can do many things these days. I hear you can play videogames on a Tesla.

Yeah... because they have computers inside them.