I guess I was understating that over the course of several runs the extra 2 credits for the multisub ice does add up. I’ll probably just stick with the Passport + Rex combination.
I think that’s a good call. Do you actually find yourself running out of Rex tokens? In my Leela play, what I’ll find is that the first decoder I draw gets installed. If it’s passport, great! Hold Rexs in hand and wait for the remote play (which against non-expert players will come too early for the corp since they’re going to think they’ve got a window for you to face check, special order, install and run again, so they’ll do it before they think you have enough cash rather than assume you’ve got one in hand). If you can’t find Rex before the remote play, then hammer a central for a random agenda and bounce the remote agenda or the ice you suspect is a code gate. If you draw Rex instead of Passport, that’s also fine. Just install Rex and use it (maybe more judiciously than you’d use Passport) and then install Passport when you find it. Since you’ve got 2 Rex in the deck, you’ll often find either the second one, a Passport or a Special Order before you run out of tokens. If you have to special order for a decoder at all in the first place, I’d grab Passport and then hunt for Rex while keeping up pressure.
I find that the vast majority of the time, if you’re careful about spending Rex tokens in the early game you’ll find your Passport and then only have to spend tokens when you run the remote for scores. I don’t think I’ve ever yet actually used all 8 of my Rex tokens in a game, since if a game goes on that long it’s probably a grinder that has slowed me down on remotes anyway.
I haven’t had a problem with running out of Rex tokens in games that I’ve played, but in reviewing the winning corp decklists on stimhack, it seems a lot of decks are going more code gate heavy than what I’ve played against in the past. I guess the big thing is just trying to keep up enough central pressure that it doesn’t get to a point where the corp can stack code gates on a remote.
Yup, that’s my experience. 4 tokens should be enough to get into a remote once when they drop something in there you need to nab, and if they need to spend a lot of time building up the remote to lock you out then they’ll have let the agenda density in centrals pile up to where you should, with multiaccess, be able to score -something- to bounce the piece you need to remove from the remote 
I used to play a Yog0saurus Kit deck and that, too, has the built in possibility of lock-out. What you find in netrunner is that so long as your possibility of lock out is hard for the corp to achieve, you’ll rarely need to worry about it as a threat.
I’m a big defender of Peacock, I think it’s much better than people give it credit for.
As I see it, it’s not a card for aggressive decks or decks that want to make as many runs per turn as possible. For example, it’s a terrible choice for the typical Gabe deck and is far better suited for controllish builds heralded by Stirling and Andromeda that focus on well-timed runs and R&D locking.
Basically, the question is, how does your deck react to Quandary? Is it a bother because it keeps you out for one more turn or is it laughable because it doesn’t tax you enough? That, for me, is the big difference between Peacock and Zu.13.