§87

Words like 'red', 'dark', and 'sweet' are not 'indefinables' (the 'simple ideas' of the classical empiricists).

Re: Moses: In §79 Wittgenstein suggested that "Moses did not exist" may mean various things (depending on which definite descriptions we 'lean upon'). But in §87 Wittgenstein suggests that I might specify what I mean (to remove doubts about what I mean) - I could say, "I take 'Moses' to mean the man, if there was such a man, who led the Israelites out of Egypt, whatever he was called then and whatever else he might have done". But, having removed this doubt, further explanations might be needed (about what 'Egypt' means, about who the Israelites are, and so on). Does this mean explanations are to go on forever to remove all doubts? - No. Explanations could be used to remove misunderstandings but we need not go on giving explanations unless we require one to avoid misunderstanding. - We don't need explanations to remove every misunderstanding imaginable.

Explanation of meaning = rule for the use (a use) of a word.

Wittgenstein is not a sceptic. Something like a signpost or a rule is perfectly in order ('complete') as long as it fulfils its purpose.