W10 seems a bit passive – we have one pro game for this position, where White played D10 (http://ps.waltheri.net/database/game/77647/). I generally find that the weaker you are the harder it is to use the split.
W12 is a semi-joseki popular in the 19th century. It’s said to be a bit worse for White locally if Black invades at C3 (see Waltheri for vars.) Then again, it’s a little closer to the undefended wall than the normal response so that could be relevant.
W16 is usually at E3 – eg. E3, D2 E4, E2 B3. This sequence is more critical of Black’s shape, encouraging him to play a gote defence at H3 at some point. If White wants sente he can use B4 – eg. B4, B3 C4, E2. I think this would be fairly strong considering White’s longer C7 extension.
B17: I think the “honte” move is E2 to make a more defined eyeshape.
B19 is arguably a little slow since Black has a forcing move at B2 for more eyespace, and can still jump into the centre.
It’s hard to say how best to respond to White’s 3-3 invasion of the lower right. I think there are a number of different reasonable strategies.
W38 is overplay. I would recommend tenuki’ing either to invade trhe left side (eg. D10) , build the bottom side (eg. E6), or strengthen the corner with G17.
B53 is slow – attacking the left (eg. C13) or the right (eg. R12) are big.There follow a few more slow moves by both players.
B63 is probably a little too close – see previous comment.
At B73 you have to first atari at B13 to prevent White connecting.
W84 is sound. B85, though, allows White to first descend to B18 – if Black stubbornly blocks at A17 then White can cut at D18.
The endgame is not that interesting to me.