That is what I was thinking. There is a planned short-turn on the northbound Spadina at the current Downsview Station. The Sheppard line would feed into the Spadina line north of this station and associated cross-overs. There is also a fair bit of space on the Sheppard Line west of the existing Downsview Station to store trains, if needed, so they can be slotted into the Spadina line at the correct time to make sure the frequency of service is optimized on Spadina line.
As you said earlier, this way the Sheppard subway goes from York U (or Steeles West - not sure exactly where the next set of cross-over tracks are) to Don Mills (or Agincourt GO). This northern crosstown line can be accessed from the Sheppard East bus, Finch West bus (or LRT) and maybe the Steeles West bus.
Yup exactly. Downsview isn't really a 'must-hit' in my opinion. Downsview Park is a bigger one, because of the connection to the GO line.
Slotting Sheppard trains in going northbound would be more of a challenge than southbound trains, but given the fact that some Spadina trains are stopping at Downsview, the Sheppard trains can fit into that same time slot as the Spadina train that's stopping. Shouldn't be too big of an issue until the frequencies get much higher.
Sheppard West has some merit for connectivity sake and operational improvements (service disruptions, carhouse movements), but still ranks low on my list of new subways needed.
I would definitely agree with this. Here's where I would rank it:
1) DRL (interlined with the University-Spadina Subway, Union to Pape)
2) Yonge Subway ext (as part of the DRL reconfiguration, King to a new Union platform or Southcore)
3) Bloor-Danforth (Kennedy to Sheppard & McCowan, assuming the SRT replacement can be changed)
4) DRL (Pape to Don Mills & Eglinton)
5) North Yonge ext (Finch to RHC)
6) Sheppard West ext (Yonge to Downsview, interlined with Spadina to York U)
7) Sheppard East ext (Don Mills to Victoria Park)
8) West Yonge ext (either Union or Southcore to Dufferin Station via the Ex)
As you can see, pretty far down the list, but one of the few projects that can be done for under $2 billion. That would likely make funds more available (especially if they're taken from the SELRT), and thus would probably move the project up the list slightly.