The point about censorship is an important one. If the system is build focused on a central hub this can be prevented directly in the smart contract. If the system is more general and everyone could potentially be a central hub (note that any store could what to have a node to receive payments, or small communities would also be interested in having small central hubs, etc) then it would be difficult to control censorship.
All the other points described by jtremback are a bit forced. In particular I do not understand the payment processor part, the central hub do not receive funds from anyone, these will be in the smart contract of the main system.People are just storing funds in a contract, those funds are realesed when the users show a signed agreement between them, is there something I am missing?