Single-argument doesn't capture the difference between functor and applicative.

  ex1 :: Functor f => f (Int, Int) -> f Int
  ex1 = map (uncurry (+))

Applies a multi-argument function within a functor. What applicative allows you do is take a product of contexts to a context of products, and lift a value to a context.

  pure :: a -> f a
  zip :: (Applicative f) => (f a, f b) -> f (a, b)

This can then be used to make ex1 into a multi-argument function of contexts, which is not the same as (uncurry (+)) being a multi-argument function

  ex2 :: (Functor f) => (f Int, f Int) -> f Int
  ex2 = ex1 . zip

Sure, it's a particular interpretation of `(f a, f b) -> f (a, b)`

    (a, b) is an a -> b
    (f a, f b) is an f a -> f b

Applicative, over an above this, is really just useful when you have (f a, f b, f c, ..)

You're just reducing my idea to a specific case. You missed the generalized idea.