My opinions on the lot:
1. Electrification was thought to require grade separation of all intermediate level crossings, but the Sunbury project thankfully provides a precedant against that. Sure, grade separation is nice, but it is expensive and not always necessary.
1A. However, if electrification were to occur, I'd prefer to wire as far as Somerville rather than Baxter, because Somerville has a greater local population (I'm told) and because the only reason to run wires to Baxter instead (other than the sidings and the general cost cutting) is to allow for a future Mornington Line restoration. I've run the figures and worked out that a train to Mornington East via Baxter would take 15 minutes each way to/from Frankston, but a bus running north/south along Nepean Highway does the same job in only a slightly longer running time, serves a lot more people along the way, costs close to nothing to create because it's already there and costs a lot less to upgrade to a more frequenct service.
2. Restoring Langwarrin station in its original position would be an absolute waste. Move it south, to Robinsons Road, if at all. But fix the bloody local buses first, in particular the 772-776 range. They're pathetic. In fact, that applies to most buses on the Mornington Peninsula.
3. Frequency on the line cannot be increased currently, because the signalling system does not allow for it (although the former electric staff and train staff and ticket system did). The easiest solution is to add two signals at (say) Somerville, one in each direction, and have that as a dividing section.
4. If any part of the line is to have a passing loop incorporated, my preference is full duplication from up of Tyabb to long Island Junction. However, it should be done in such a way that both lines are bidirectional, so that it can be used as both a passing loop and as a holding track for the Long Island freight trains (assuming they stay on broad gauge after the E-Gate residential project abolishes the steel train gauge transfer sidings).