Even though Finley wasn't used in that fashion, defenses respected his ability to be used in that fashion, which helped open up the field for the receivers. Right now the team doesn't have a player that can threaten the seam and force the safeties to respect that threat. Now, normally that isn't a problem because most defenses aren't good enough to take away the receivers, but we've now seen the lack of that TE threat against Minnesota, New Orleans and now the Bills.
Perhaps if the team was more willing to run the ball consistently but I'm pretty sure D corrdinators don't actually fear the run as much as they should because they know there's a pretty good chance the Packers won't stick with it.
Finley, like Graham before him, tried to get himself classified as a WR for franchise tag purposes as a negotiating ploy. There was some merit to his claim given he ran about half his snaps from slot or wide out. Like you say, he'd run a lot of intermediate-to-deep seam routes out of the slot to either draw safety coverage away from the WRs or provide an attractive target if teams tried to cover him with a LB or nickel back.
That does not speak to possession routes.
Cobb out of the slot serves the same purpose as Finley but in a different way. Instead of having a jump ball seam TE, he runs a lot of crossing routes in the intermediate zones. As I posted in another thread, Cobb was the league leader in first downs per target going into week 15, by a wide margin, among the top 30 leaders in yards receiving.
The formulation is the same...opponents can't double two WR and a slot. Against New England, Adams was left one-on-one and the match up was exploited. Buffalo was mixing it up. Cobb had a lot of opportunities; we saw Nelson wide open in single coverage on the dropped pass, Adams had single coverage opportunities but was not connecting with Rodgers on the route.
To me the problem in the Buffalo game (besides the drops and misfires), and the fact that Buffalo has an awfully good defense, was that the Packers did not run the ball enough.
Buffalo came out playing what they played against Denver...nickel, 6 in the box even on run downs, occasionally bringing down a DB over the TE. The Packers had early success running Lacy, but went away from it.
The issues in this game were pass game execution and philosophy (run to set up the pass; short pass to set up the deep pass; don't turn the ball over).
While having a big, semi-fast TE with hops would be a nice weapon to have again, there are different advantages to having a WR in his place in the typical 3-wide set.
In the end, this is a big play offense. As much as I'd like to see a switch to control-the-ball mode in mid-game when the situation dictates, it's unrealistic to expect a 40-45 run/20-25 pass mix the way Denver has been going in recent weeks. The Packers just don't have the kind of defense to be relied upon in close, low-scoring games. That said, Lacy + Starks running the ball only 19 times looks a bit stubborn under the circumstances.