You can increase efficiency a bit by manipulating path directions. You’re unlikely to save a lot of time though.
You’ll typically spend more time modifying paths than you’ll gain through a burn or two or three. If it’s a job you’re doing repeatedly, it can be worth it.
As for the order, I don’t know if anyone’s played around a whole lot with manipulating the pathing of the motion planner. Not that I can recall. The order can be somewhat of a mystery.
Dan has said before they haven’t spent a whole lot of time on pathing optimization because it takes a lot of effort for minimal return in overall job length.