Just because the amount of work to complete it is _now_ relatively small doesn't mean that this has always been the case. The command line version may well have been available for some length of time but this is no indication that a full GUI version (with an appropriate plugin interface) was in any way possible two years ago (in fact, I can pretty much guarantee that it wasn't). The work done over the last two years has done a number of things:
a) Improved UnixLib a very great deal - just look at the changelog to see what's been changed.
b) Improved the porting tools which facilitates future ports.
c) Increased the number of libraries available for such ports to link against.
d) Improved the X compatibility library (ChoX11) to enable ports to run natively under RISC OS without requiring an X server.
Note that points a and c directly benefit native applications, too.
Obviously, most of this has no impact upon the end user so it is very easy to say that no progress has been made. However, without such developments, you can pretty well forget any possibility of improved (or new) ports appearing. Note that each and every new port undertaken uncovers issues with various supporting libraries and parts of the toolchain, thus diverting effort into fixing these issues first. Is this a waste of time? No, as it reduces the effort required to make the next port. However, this all takes time which, with a small number of developers, is an extremely limited resource.