I honestly think only someone foolish or ignorant could think that creating a better loglang than Lojban is not trivially easy. You just put phonological form to a Polish or Reverse-Polish version of predicate logic notation. Doing better than Lojban is really not a meaningful benchmark.
Well, this is the core of Xorban, of course; it takes care of the log part. The interesting question is about doing the lang part. So far, Xorban has set out a few moves -- not all fully realized yet, naturally: speech act indicators (so that there are other than declarative sentences), interjections (so that there are other than sentences, though sentence fragments are not yet discussed), expanding possible predicates to include at least events as arguments (presumably propositions are in the offing), some moves toward sentence melding, introducing as shorthand for several sorts of constants (starting from a term-free version of FOPL), considering othe logical (though not standard) operators (tense, aspect, mode, maybe mood). In all of these, it is following Lojban, though, hopefully, with less complexity for the same expressive power (time will tell).
--And.