I largely agree, but a lot of these are minor considerations compared to the four BIG changes that deserve top priority: Food Preservation, Realistic Recipes, Fermentation Time, and of course Farm Size.
Food stockpiles were always merely a placeholder for something meaningful; you can't protect your food from rotting by simply telling the floor that you've designated it as a food-storage area. There really needs to be at least one new labor, probably requiring its own workshop, of at least one of the following: Drying, Salting, Pickling, Curing, Smoking, Refrigeration, and Canning. Of these, Canning is the only process not practiced in medieval (and older) times, and when you get down to it, there's no real reason why it couldn't have been. (It was just a quirk of Earth's worldgen that put the only rubber trees on the "wrong" continent, and rubber isn't actually required for canning anyway--the jars could be sealed with wax or pitch or something instead.) Different civilizations could start out knowing different preservation methods (for reasons of replay flavor), but every civ does need to know at least one method.
The container a food item is placed in will determine its vulnerability to vermin. Its preservation method (and perhaps the skill with which it was preserved) should determine its rate of spoilage (perhaps with a random +/- 10% modifier as well).
As a side-suggestion, maybe there should be a new minor noble: the Panter, whose job is to control & maintain the fort's pantries. He must know the quantity & location of all food items made in the fort--not necessarily on the entire map, just those produced by a Kitchen or Still, or preserved by one of the methods above. He might not know about the meat from animals/fish that have just been butchered, or plants that have just been gathered, but he makes regular inspections of all food stockpiles (especially around harvest times) and adjusts his tally accordingly. He functions both as a subordinate Bookkeeper (ideally, reporting to the bookkeeper once a month in her office, where they Update Stockpile Records together for a few moments) and a Manager, as he has the power to automatically queue up Preservation jobs, or Cooking jobs, (always one or the other, with Brewing jobs counting as Preservation) to be performed on all fresh foods (except those that keep well, such as grain). The Panter is also in charge of knowing how soon each container of food is expected to spoil, and has the power to rotate stock by automatically "soft-forbidding" containers of food as soon as they're preserved, and un-forbidding them once they're either a) nearing their expected expiration date, or b) among the oldest food in the fort.
The existing biscuits, stews, and roasts were also never anything more than placeholders, an early way to include some suggestion of food variety and player control. It's difficult to discuss "Units of food eat per job" until we define what a "unit" of food actually is. In a plate of, let's say spaghetti & meatballs, is the amount of Parmesan cheese supposed to be equivalent to the amount of spaghetti, or tomato sauce, or meatballs? Or are they all present in different amounts? It's fine to say a "unit" of food is equivalent to one dwarf-meal, but if I go into my kitchen and combine [1]bread, [1]lettuce, [1]ham, and [1]cheese into a sandwich, is that sandwich going to sustain me for four meals, or feed four people? No. There needs to be a major overhaul in how the amounts of food are represented to the player, and especially how they're handled internally. I'm fine with prepared foods & drinks being expressed in terms of how many dwarf-meals they represent, but the Kitchen needs to be able to handle quantities that are much, MUCH smaller than a full meal. And the best way to do that, is to implement realistic recipes, which would make the game FAR more satisfying anyway.
Not only should brewing beer and fermenting wine take months, or even years, to yield a potent (or even appealing) batch of booze, but they are also very different processes that fully deserve to have their own separate workshops, and separate skills. A Brewer is not the same as a Vintner is not the same as a Distiller, and a Brewery is not the same as a Winery is not the same as a Distillery. Beer and wine (and cheese as well) need to sit for a LONG time before they are ready for consumption. A Distiller can work quickly, but they need wine as a starting ingredient, and the quantity of output is always noticeably smaller than the input.
Last but certainly not least, it's flatly ridiculous that, with a Legendary Planter and fertilizer, a single farm tile is able to more than satisfy the food and drink requirements of TEN DWARVES, perpetually. It gets even sillier when you factor in that a great deal (most?) of these tiles do not get their energy from either sunlight or any vague "cavern magic", they are merely regular old dirt in any common soil layer. And yet they apparently contain a literally explosive and limitless amount of chemical energy, which all dwarven crops can metabolize. I have never heard of any player who thought that farming was either a) too difficult, or b) too unproductive--quite the opposite, in fact. I know that traditionally, cavern-dwelling dwarves don't have access to the acres & acres of rolling farmlands that humans do, so their farmers have to do more with less . . . but for a single farmer to be able to feed a maximum-population city off of a farm no bigger than a Trade Depot is just beyond the pale.