I think the whole idea of this type of rubato pertains to, and originates from, singing. Chopin's loved Italian bel canto above all, and he just wanted to emulate it on the piano. Whether it was more wishful thinking than reality, I guess we'll never know. Even with half a ton of quotes we're nowhere nearer a definitive answer.
I think we can definitely say it was reality for Chopin, if not for everyone or even most of us. The reason I say this is the matter of the vehemence on the part of Chopin's pupils and contemporaries in decrying misguided and even vulgar attempts to recreate what Chopin did effortlessly. The only prominent individuals who seem to disagree are Berlioz and perhaps Meyerbeer (Mendelssohn seems a bit wishy-washy to me on the subject) - it's not clear how much distinction should be drawn between the rhythmic irregularities in the mazurkas and the concept of 'stolen time', so it's difficult to say whether Meyerbeer had similar feelings about Chopin's playing in other contexts. Berlioz is easy to understand, mostly because he was primarily an orchestrator, and a conductor. In that context, only the soloist can have any freedom, and individual deviations from the tempo are usually associated with inferior ensembles. Aside from that, Berlioz seems to have been fonder of Chopin in the early days, perhaps before he became aware that Chopin wasn't very fond of him.
For sure, what would feel natural to any singer (they all do it, from crooners to pop singers) requires almost superhuman effort from a single player.
I don't think it's necessarily superhuman, any more than being able to memorize music is superhuman. Some people will have a knack for it, and some won't. Like I said (and like Eigeldinger said), it's probably no coincidence that Chopin chose to be represented in the Fétis/Moscheles method by polyrhythmic etudes. Moscheles probably asked Chopin for something that would help players achieve the necessary independence of the hands, or perhaps it was something simply understood between them after their meeting and performance for the royals together (which Moscheles described in some detail - the above quote is an excerpt from that). Obsessing over where exactly each note falls between the other in polyrhythmic etudes will not achieve that independence of the hands, especially when rubato - passionate declamation - is required to make it convincing. Neither will unsteady renderings of either rhythmic figure be convincing. Yes, it's difficult, but whether or not we can execute it, we can conceive of it, and perhaps aspire to it if we are so inclined.