The problem is that many organs develop through cell interactions that are coordinated in the development of the organism. One cell type arises from an interaction and then it induces other cell types to arise which then induce other cell types, etc. We can mimic this to an extent by supplying the proper signaling factors in vitro, but for a complex organ it can be tricky to get just right and keep it all in working order. Imagine a situation where you 3D print a lump of flesh that is grafted on without any nerves, might as well be a chunk of silicon. Definitely something we will see within the century, but I believe we are still a ways off at the moment.