It's not that simple. Yes, it means you have more data stored in VM than you have physical RAM. But, some of that may be e.g. cached files, which may never be paged in again (eventually being dropped after swap fills up with such "let's keep it in case we need it later" data). Even if it is paged in later, this would be substituting a slow swap operation for an even slower filesystem operation, not the harmful page thrashing when not all active processes fit in physical RAM at once.

It's IMO unfair to call this "ran out of physical RAM", and your conclusion is plain wrong -- it simply won't lead to excessive paging if the several active processes are fitting comfortably in RAM, and the only things paged out to swap are cache.

(And yeah, I'd like more RAM, and I get that swap is slow -- my point is simply that you are misunderstanding or distorting the actual behavior of the system.)