Nice camera!
You're good enough with your cross in terms of thinking through it and coming up with a full solution that you can learn a lot very quickly from
jarcs. Just generate a scramble there do it on your cube, figure out your own cross, then click the Cross tab and "solve" on the left. Compare your solution to the ones jarcs found and learn some new tricks.
First solve, third pair: The orange/green edge at UF has orange on the top, and your F color matches (meaning is the same or opposite color), therefore it's solvable with just L/U/R moves. In other words it's correctly oriented and does not require cube rotations or F/B moves to solve. You did two rotations, when you could have just done L' U' L U' L' U' L.
First solve, fourth pair: Again, the top color of the edge matches your F color in that they're opposite colors, so you shouldn't need a rotation for it. You can look at the F2L link above for ways of solving this case without rotations, but the way I would have done it is U' R2 U2 R U R' U R2.
Second solve, when you talk about doing the cube rotation: During look ahead, you should be looking both for corner stickers of your cross color, and UL/UF/UR/UB/FL/FB edge stickers that match (same/opposite) your front color. Those are the edges of the pairs you should do first, because they will not require cube rotations. Mastering this extra bit of information during a speed solve was the main thing that got me sub-10 from ~12. I noticed right away when you inserted the second pair that you had two F2L edges on the top, UF and UB, that were the wrong orientation. I would have done a cube rotation right away even if I had no idea where the corners were, because it's the edges that dictate the rotations, and ended up solving the remaining pairs the same way you did, but without having to look ahead so much all at once.
Third solve cross: You could have done R' instead of R when moving the red to the back, which would have allowed for F R L2 to finish.
Third solve immediately after cross: Again, look for stickers that match F on top and at FL and FR. You have three that match on top UF, UR, and UB, and the FU and RU stickers are not yellow meaning they are F2L edges. You could have solved those two first without needing to do rotations at all. Also notice that the LF and LB edges are F2L edges (I don't expect all this recognition instantly during a speed solve, I'm just pointing it out for learning) are incorrectly oriented. That is, the FL and BL stickers do not match F.
The issue here is that when you have an F2L edge on the top, doing a cube rotation will toggle it's orientation. If the edge is in the E layer, doing a cube rotation does not affect it's orientation. If it's correctly oriented, it will stay that way, allowing you to do a rotation for another pair and not worry about having to rotation back to solve that edge. If it's incorrectly oriented, it will stay that way, meaning you should try to get it to the top during the insertion of another pair before rotating. I would have started this solve with L' U' L to create the blue/red pair at the back left, which also brings the FL edge to the top. Insert with R' U R, noticing the orange/green pair at front right. Now, I could do R (or R2 follow-through from last pair) U' R' U' L' U' L, but because I know that the BL edge is bad, I want to remove it. So, after inserting the red/blue pair, I instead do U2' L U' L' U L' U' L'. Now both remaining F2L edges are on top, and I can do one cube rotation (or d move) to fix both.
Next I made the blue/orange pair with d R' U2' R, and noticed the case forming at the back right. Just from experience I know that when I have a state like this I want to do U R' U2' R, and that will end up with a better case than U2' R' U R.
Third solve, last pair: How you did it is fine, but if you are okay with F moves I recommend U R' F R F' R U' R' (That's starting with the corner at UBL as it was when you finished the third pair).
I didn't look at the last two.