This is a good example of "it depends who you ask".

In the case of riders like myself, who ride a pretty huge variety of bikes from all eras, I immediately note the differences in steering response when riding different bikes.

Now for the most part, a rider can get used to any bike, even those wildly outside of the normal range. I raced a Pedersen for over 20 years, and one not only has to adapt to the steering on these but also the physical adaptations that was also part of the bargain. This might be analogous to to a rider given a team bike to race, since one is expected to use what is provided.

The issue however might be how often that one switches bikes, and how much time (and how many rides) that they might allow for to re-adapt to each different bike.
I prefer steering that is neutral to the extent that I can follow a swerving paceline as needed without over- or under-reacting with my steering input.
I don't much like riding bikes that feel excessively "busy" or twitchy when descending at speed, since this seems to force me to focus excessively on the road surface instead of the trajectory of the road ahead.
And I prefer out-of-saddle steering response that doesn't require me to re-familiarize myself with steering input corrections in response to pedaling efforts, something that can be immediately annoying.
Ok, racing the Pedersen, I had to suffer through all that, just as many racers completely adapt to and endure the quirks of their racing bikes that are perhaps adapted to a most "aero" configuration (stem-wise).

But, for those of us who like to be able to ride different bikes and not have to re-adapt our steering input habits to each one, keeping track of how different stem lengths are affecting the steering might be a good idea.
I encourage every rider to experiment in this regard, as opportunities present themselves, since those experiments that we validate on the road will be the most relevant to us as individual riders.