In the interview Tolkien in Oxford JRRT replies...
'No. No. No. I wouldn't mind other people knowing it, and enjoying it, but I didn't really want to, like some people who have been equally inventive in languages [? desiring ?] to sort of make cults and have people speaking it all together, no, I don't desire to go and have an afternoon talking Elvish to chaps. For one thing of course Elvish is too complicated. I've never finished making it.'
And in any case the Neo-elvish on the web is not simply full of neologisms [not that you said so Galandir], but represents an artificial selecting of material from various stages or phases of the invention process, or as Carl Hostetter [of the Elvish Linguistic Fellowship] briefly describes them: 'artificial, simplified, patch-work systems'
Carl also writes:
'The fact is that, to the extent that we can speak accurately of Quenya and Sindarin as single entities at all, it is only as continuities of change over time, i.e. as processes; all else is simply individual snapshots of (most often only small parts of) this process, any detail of which may have persisted from the beginning to the end of that process, or have had no more extent in that process than the sheet it was written on; and in some cases there may be no way to tell which of these two extremes is true of any given detail.' from the FAQ over at E.L.F.
Some may disagree with that... I however agree
Plus we must remember that so much of what Tolkien put to paper was not only subject to change, but was never put into a 'finalized' form meant for readers -- publication. Etymologies, for one example, provides a lot of 'the corpus' -- but what it really is is an abandoned document, a document written before major historic changes were to take place...
... and here I mean Tolkien drastically changed the history of his languages and put Etymologies aside, so that the relationships of various Elvish tongues [to each other] in Etymologies was no longer true, never mind whether or not individual words or grammatical details would necessarily be carried forward when 'Noldorin' essentially became Sindarin, for example.
If we think of the amount of Elvish Tolkien actually published it is relatively small, and JRRT could not resist changing some of that [already published material] between editions! That doesn't mean I toss aside everything else he wrote, certainly not, but what I do often enough see, or seem to see anyway, is Tolkien 'inventing' on the spot and enjoying it. And as Carl Hostetter also underlines, Tolkien not only never 'finished' any of his invented languages...
... he doesn't appear to really be trying to finish any of them, as in, he didn't desire to make them finished enough for others, or himself, to learn how to speak.