That's an interesting list of potential options, but I guess I see substantially greater downsides in production, lower expected production, and a high prospect cost in any of the trades.
The Cubans are much more likely to be JC Linares or Dayan Viciedo than Yoenis Cespedes or Abreu, yet they'll cost as much as a mid-tier free agent.
The most attractive mid-tier free agents you listed are PED guys, I just don't buy that Nelson Cruz is doing what he's doing as a clean ballplayer. He wasn't ever caught by the testing, so he's probably still doing whatever it was he was doing in the first place, but with a more discrete supplier this time. Once he gets a long-term contract, he'll stop using the drugs and go back to being a guy with average power and below average plate disciple, essentially Jonny Gomes. Coby Rasmus is a total tease, whose inconsistency would be maddening on a 3 year $39 million commitment or perhaps even higher as some team still pays for promise rather than performance. Aoki can't hit, Cuddyer, Cargo, and Dickerson come with the Coors factor, and the reason the Orioles are unlikely to pick up Markakis' option would be that he was awful for 2 years prior to this one and they would know more than others how much weight to put on each in calculating the the three-year average production.
I don't see the Cards as willing to trade Craig for what you're suggesting; they'd probably want at least Marrero, who'd allow them to move Peralta to 3B and Carpenter back to 2B, and one of the righthanded pitchers from AAA.
I still say Ellsbury or Choo was where to spend the money in the outfield for the next 6 years, and then fill in around that with the propsects the Sox have in their own system. To me, at this point, the best option is to keep the Nava/Gomes platoon while you figure out who among Betts, Middlebrooks, and Cecchini can play the best outfield corners, and whether Bradley can hit.