An interesting topic - and you are right it is not an original idea - there are several examples of this approach. When applied to the real world it is clear that FRFFs may not be the panacea we would hope they would be.
First problem is siting of the freshwater rearing site. If they are placed on a large river system we would likley have adverse impacts on existing wild stocks. If they are place on a small stream without exist or miminal salmonid populations the behavior of the return adults will be such that they will not enter the stream until they reach maturity.
They will mill around the mouth of the small stream until they are ready to spawn. The fish returning to the "processing plant" will be dark and of reduced qualtiy. For the process to achieve a high quality product they will have to fishing in the terminal area creating the same problems of mixed stock fishing and fishing gear impacts. Additional such a scenario results in same FRFFs straying to the near by rivers spawning with existing wild stocks. The situation of your FRFFs is almost exactly what the Tulalip tribes have created here in Puget Sound. Their hatchery is a small stream with minimal wild salmonids (mostly coastal cutthroat). However they can not get enough chinook or coho back to maintian their program, rather they get eggs from a state run hatchery. They are successful in getting chum salmon back to the hatchery. The fishing on the return fish occurs in the terminal bay as well as the outside mixed stock area. Generally the fish caught in the terminal bay are of poorer quality.
Their chinook program illustrates the problem with FRFF produced fish straying to nearby systems. Roughtly 10% of the natural spawning chinook in the nearby Snohomish system are from that program.
You are correct in that FRFFs would end the sea-lice problem - a big plus. However the release of the smolts create both predation and competition problems for nearby wild stocks - trading one problem for another.
Much has been made of the use of fish meal for feeding of farmed stocks and the potential over-mining of forage stocks. This is true but the real issue is we as humans insist on feeding high up on the food chain. We want our fish protein to be salmon rather than herring fish meal. It is pretty easy to demostrate that if we are going to eat X tons of salmon the farmed product makes a much more efficient use of the forage base than wild fish. Higher food conversion rates and less mortality going to factors other than human consumation.
FRFFs would be unlikely to produce the year-round availability of product that the farmed fish do. You are correct that there are year -round runs of various salmonids. However most still retain relatively narrow spawning windows which limits their availability in the home stream unless they are in large river systems. For example while steelhead of one stock or another return to home systems every month their spawning still is confined to late winter/spring/early summer. When they are produced in a small system they are hestiate to enter them until sexaully mature thus making them only seasonal available unless their harvest is moved into the larger (mixed stock area) streams or marine waters.
In short except for some very special and limited circumstances FRFFs present as many economic and biological problems as either net pens or commerical fishing on existing hatchery and wild stocks. FRFFs should be consider just one of the many tools available to supply protein for our diets. That tool is likely only the best one in limited cases.
Sorry to be negative however if we are really going to address the problems generated by society needs we need honest assessment of the resulting problems and realistic solutions based on considered and informed risk assessment judgements- there are no magic bullets to solve the problems.