The way this article is written makes it sound like "dense star forming regions" somehow form stars with planetary disks in isolation from each other, which then get close enough together to have their disks interact to form these binary pairs of matter.
What seems unlikely to me is that the stars would form and have protoplanetary disks that did not interact while they were in their formation processes, and then "collide" later. I would expect their disks to form with some varying levels of interaction from the beginning.
The other thing about the article which seems odd is the statement ""Planetary mass objects may represent a third class of objects, born not from the raw material of star-forming clouds or via planet-building processes, but rather from the gravitational chaos of disk collisions."
First, it is the same "raw materials" for everything that gets formed, including the stars, their planets, and whatever these JuMBOs are. It is the process of the development of the high density regions that will collapse into the various type of objects that may have some differences.
So, I am wondering what differences would be expected in the resulting JuMBOs from "Jupiters" and "brown dwarfs". Can we make predictions and observations that would show those differences?
And, I am wondering if the reported decreasing probabilities of being in binary pairs for stars of decreasing sizes could be mainly the result of bias due to the decreased ability to detect the smaller, non-luminous, not-very-reflective objects. "Binaries" do not seem to need to be "equals" to form, but perhaps are easier to detect when they are nearly equal. I do not know of any reason to expect equal mass binaries to form with greater probability than unequal mass binaries, other than our naming conventions. Would we consider a red giant with a brown dwarf satellite to be a binary?