Okay, I have an awesome head-cold, so it feels a little like my brain is floating in yoghurt. Anyway.
Since you tried to veer away from the specifics I'll try to give a more generic answer although I hinted at it in the previous post:
The dominant factor when I decide whether to call or re-raise with a flushdraw on the flop is stack sizes. This is hardly surprising since stack sizes are the dominant factor in everything I do.
After that, though, comes position. Out of position, I almost never like calling. If I call, it's essentially a crying call because I realize that my opponent's range is such that if he raises me, he has me in really bad shape and re-raising or shoving will only get more money in bad.
In position, I'm not a huge fan of the "wohoo, flush draw, allin" philosophy that many of the regulars I play seem to have adopted. When I have positional advantage I want to make sure I make the most of it. Getting it in on the flop means the hand ends there, not one or two streets later. That benefits my opponent. Whatever benefits my opponent is detrimental to me. That said, there are certainly times I'll shove with a flushdraw in position on the flop. It's just not quite as appealing as it is out of position.
The deeper the stack sizes, the less willing you should be to get it in with a flushdraw, obviously. But what also matters is that you get the last bet. This is overlooked by many, even by many otherwise solid players. Your hand is an awesome semibluff. In order to leverage it completely, you need to be the one who shoves and your shove has to be such that it
1) can work as a bluff (i.e. not shove your remaining $10 into a $160 pot), and
2) Isn't so big of a raise that the cost of the bluff makes it unprofitable.
Here, you're borderline but (probably) still profitable. If the effective stack had been $130 or so, I'd have been very weary of shoving. When called, you will usually (since you don't have the nut flush draw) be only around 35% to win on average, assuming his calling range doesn't include anything worse than what you have.
Assuming you started with 100BB: When called, you will lose 0.35*(100+6+4) - 0.65*(-93) = 38.5 - 60.45 = -$22.
If there's $35 in the pot, your bluff needs to work 22/35 ~ 63% of the time. NOT TRUE SEE EDIT. In other words, he needs to have something less than a set/two pair/TPTK that he can't let go about 2/3rds of the time. This is a little much to ask for, and the "problem" in this hand is that there was no raise preflop so the stack-to-pot ratio is high. Still, if you think his range is wide enough to include other things than sets (e.g. straight draws, nines, QJ, A8, AJ taking a stab, etc) then it's usually pretty easy to get enough combos to make it profitable, even this "deep." Only 9 combos of sets, after all.
Edit: Yeah, so, I have a cold. And I'm sitting here after I posted that thinking "22 over 35. That seemed easy. Is it really that easy? Why did I think it was 22/35?"
And then I thought to myself, "no, that's not true at all." and then I wrote down the actual formula for EV and what we get is this:
If he calls our shove X% of the time, then we will lose $22*X%, and we will therefore pick up the dead money (1-X)%*35$.
Our breakeven point is therefore -22x + (1 - x)*35 = 0
x = 61%.
I.e. he only needs to fold 39% of the time for us to break-even. I thought 63% sounded high.
End edit. Let me know if I did this wrong, too; hopefully my head will be back to normal in the morning. I'm going to bed.