The more I think about it, the more I come to realize that creating a new strip of land immediately in front of the current waterfront makes sense on a lot of levels.
1) Presumably less expensive than having to tunnel underneath an existing arterial road and existing elevated expressway (the Central Artery Expressway scenario in Boston).
2) Significantly less disruptive (for the reason mentioned above).
3) At the end of the day, Toronto gets acres of new valuable real estate from the removal of the Gardiner, AND the opportunity to create kind of a Grant Park type of deal right along the waterfront, something Toronto is sorely lacking.
The way I see it is you build a new concrete seawall further out into the harbour (30-40m), with the wall extending down maybe 20-30m. You then drain the basin that is created, dig it out, and create what is in essence a giant concrete tub along pretty much the entire central waterfront. You then build the tunnel and all the associated infrastructure in the tub, and then when construction is complete you fill in the rest of the tub with excavated fill from the DRL or another subway project underway.
This would allow the entire construction to take place in relative isolation, with the only significant impact being to harbour boat traffic and residents along the waterfront (who would be disturbed anyway if the tunnel was dug underneath the Gardiner).
To smooth traffic flow to and from the highway, I would maybe suggest turning Simcoe & York, and Bay & Yonge into alternating one-ways in order to minimize the number of on and off ramps, similar to how the surface streets feed off of the Queensway in Ottawa (all N-S arterials in downtown with the exception of Bank and Elgin are alternating one-ways, with uni-directional ramps onto the Queensway). The biggest reconfiguration required would be at Front and University, because SB traffic would need to be diverted primarily onto Wellington and then onto Simcoe, but that's doable.
All and all, I think this type of approach offers many of the benefits of the tunnel concept, without a lot of the negatives (mainly construction disruption).
Just an alternative to consider.