Fascinating article, but I think it could benefit from some more examples. Since it mentions "a tree evolving into a dandelion (or vice-versa)" and then a paper about wood evolving independently at least 38 times in the Canary Islands, the perfect example is Sonchus Canariensis (https://www.smgrowers.com/products/plants/plantdisplay.asp?p..., better picture: https://garden.org/pics/2014-03-30/Kelli/5ef4db.jpg). I saw this plant during a hike on Tenerife a few years ago and found it so fascinating that it has stuck in my mind ever since. It's basically a colony of dandelions (ok, they're a bit larger, but still clearly dandelion-like) growing on long woody stalks. Not quite a tree yet, but definitely a shrub. I guess its peculiarity of only having leaves at the tip of its "branches" limits its growth, because the water circulation in a "tree" is made possible by the negative pressure produced by water evaporation on the leaves, so a "branch" can only get so long with leaves only at its tip?
And, since it later also mentions theoretical "insect birds"... well, there are some insects who are regularly mistaken for hummingbirds: https://en.wikipedia.org/wiki/Hummingbird_hawk-moth. In photos you can quite clearly see that it's an insect, but if you see it in real life it hovers around flowers just like a miniature hummingbird, and it's a bit large and bulky for a flying insect, so the confusion is understandable. BTW this is also an example of convergent evolution, and it's interesting that these moths don't exist on the American continent, where the actual hummingbirds have occupied this ecological niche.