I am not sure that I would recommend struggling on with this, if you still are. It is in one sense as ghastly as some have suggested. When I managed to solve it, it was with more of a groan than the warm glow of satisfaction that I experienced after finishing the recent puzzles by Shackleton and Samuel, for example, when all the pennies were in place.
However, if you are determined, here is a suggestion for gaining entry to the grid. Write down the answer lengths to all the clues you have solved and look for a symmetrical pattern; that will guide you to the positions of the across clues in the bottom two-thirds. When I did this I found two answers that obviously had to be located in the centre rows of the bottom two-thirds, and there were only seven different ways of entering them. After that the down answers fitted relatively easily, two of them surprisingly non-symmetrically, against my initial expectations.
The wording of the preamble left something to be desired. A note explaining that the clues are in the usual order would have helped. And numbering the clues "a", "b" etc would have done away with the necessity of explaining what the numbers were for. I agree that Phi should have told us how many misprints there were and perhaps also how many words in the hidden work. I am normally in favour of including word lengths too, and although that would have made the puzzle easier to solve, it would have increased the pleasure enormously for many solvers. Actually, I don't see why he couldn't have given the bars in the bottom two-thirds, leaving the top third carte blanche; that would have produced a much more admirable puzzle, I think. This is another example, I think, of a setter trying too hard to raise the difficulty bar, instead of concentrating on giving solvers pleasure, which should always be the first priority.