The entry fee is the only way to do congestion pricing that is predictable and scaling. To achieve those it has to get done in advance. If the cheapest tickets have been sold in advance for the day you want, you can still get a ticket but the price went up. Meanwhile you need to know if you'll be able to get into Venice (or afford to) when booking tickets or selecting hotels.
So your plan works if you just want to raise money/raise the cost to discourage people. It doesn't work if you plan on increasing the cost to act as a soft cap on the tourists per day and you want people to know if they got in on a given day.
I don't really understand what you're saying. How does needing to book two things (transportation and entry ticket [0]) make for a more responsive market of people deciding to come/not? It seems like people would be more likely to get a deal on transportation and then pay higher congestion entry, or vice-versa.
Meanwhile, baking the entry fee into modes of transportation bundles it all in one ticket, letting people make one decision to go/no go. Each of those entry modes has a given (or at least designatable) capacity, and would have its own increasing price curve administered by the ticket seller. This has the added bonus of reserving capacity on each transportation mode for the locals traveling at the last minute.
[0] Hotel seems to bake in the entry ticket, which actually keeps the book-factor at the standard tourism "2" (transport+hotel), and only increases it to 2 for day trippers.
So your plan works if you just want to raise money/raise the cost to discourage people. It doesn't work if you plan on increasing the cost to act as a soft cap on the tourists per day and you want people to know if they got in on a given day.