Good points. I think the emphasis on natural regeneration is a case of the pendulum swinging back to counter the hype and hoopla that often accompanies big tree planting projects (the criticism surrounding the Trillion Trees project as an example). And it is true that natural regeneration does work better in many cases but much of the examples the Robin Chazdon and others cite or refer to involve degraded areas that are continuous with larger intact ecosystems (such as near Nyungwe, Kibale etc). When degraded forests/ecosystems abut such larger intact forests/ecosystems, there are opportunities for natural recovery through seed dispersal from adjoining areas etc.
The question about active restoration arises when such degraded areas are in the middle of nowhere or if seed sources or mother plants are simply not there in the landscape anymore. In our own case, we found that rainforest fragments that were more isolated responded better to active restoration, while those near larger tracts of forest less so (and could be left to themselves for more efficient use of resources).
Right now as a part of our restoration efforts, we are also experimenting with different restoration treatment methods (leave it alone, weeding only, seeding only, weeding+seeding etc) to assess what works better or how the recovery differs under different treatments.