I hesitate to respond to you because your tone seems especially combative.
It is well known that the relevant wake friction calculation ("classical Chandrasekhar") is for a pointlike structureless body. Satellite galaxies and their parents are not this. Importantly satellites can lose matter and thus mass, so one has to correct for the evaporation and tidal loss (especially to the halo) of the satellite galaxy; and the host galaxy's potential grows adiabatically over the lifetime of the system.
If the density of the halo and other sparse matter increases closer to the host galaxy, inward migration of the satellite sees two effects driven by the collision into more particles: more satellite mass is torn off by tidal interactions and more halo particles surrounding the host are swept into the satellite's wake. All that matter stays well above the surface of the host galaxy rather than descending into some inner halo or deep into the host galaxy itself. This serves to "suspend" the sinking satellite, posing a challenge for the "they should sink" reasoning; it also reduces the amplitude of the wake, which moderates the drag on the satellite. The combined result is that the evolution of specific angular momentum may not be enough to clearly support Kroupa's claim in the fine article linked at the top.
There has been a lively dialogue in academic publications about the scale of these effects for some twenty-five years. Kroupa's article touches upon his own position in these, but not those of others (e.g. HS Zhao, Binney) who work and publish in the area (with all of the above citing one another from time to time).
I don't mind this at all, personally, as at worst Kroupa is omitting something an interested reader could hunt down her- or himself, and after all it was "only" published on a general interest site whose present front page (https://iai.tv/articles) has an article by Eric "electric universe" Lerner entitled "The Big Bang Didn't Happen".