Track change is probably great if you are dealing with a multiauthor presentation document, a contract, or something of similar length. With a 140,000 word novel? Not so great. It's becoming standard practice to use it in book publishing because it is easy to follow who has made what change, allows layers of sidebar comments, and doesn't involve unwieldy piles of paper and and a set of different coloured pens, one of which is guaranteed to run out of ink halfway through. It's fast. It's kind of efficient (although its command structure sucks). But it also encourages the user to concentrate only on changes rather than the context in which they are embedded, and in a novel context is all, and in a strict sense, every sentence depends on every other sentence, because each reacts to or builds on, contradicts or enforces, its predecessor.
Track change, with its helpful marginal bars and coloured highlighting, privileges changes over the rest of the text. Which is of course sort of the point, because at its most basic editing is about pointing out glitches, omissions, inconsistencies and plain old mistakes, and making suggestions about fixing them. But changing one part of the text, sometimes even a word, can affect other parts of it - the parts you might not see while concentrating on nothing but the changes track change tracks.
Which is why I'm now reading through the entire manuscript, sentence by sentence. Partly trying to make sure that fixing problems highlighted by editing hasn't created other problems, partly polishing the text. Removing superfluous commas and adverbs, making sure sentences aren't back to front, checking that the things characters say are the kind of things they would say, so forth. It's terrifically useful to have others critique the text, but in the end, all writers should be the harshest editor of the thing that came out of their head.