Well, the "good" news is that the nuclear warheads fielded by both India and Pakistan are pretty much all in the 12-kiloton range... i.e., they are "only" Hiroshima-size atomic bombs; if used against cities (which they would be), figure each would cause 300,000 deaths -- 100k "instantly" and 200k from radiation (an estimate from the Japanese experience, which is probably comparable given Japan's high population density).
Ergo, I'm guessing that the total kill-potential of all nuclear warheads on the sub-continent is about 25 million... 7.5 million for Pakistan's arsenal of 25 atomics, and about 18 million for India's arsenal of 60 atomics. Well, we could probably quadruple that "base" number of blast and radiation deaths to account for the resulting chaos and starvation which could likely follow... call it 100 million dead (25 million Indians and 75 million Pakistanis) in the case of a full scale exchange.
Now, that is horrifying, no bones about it. However, the total nominal yield involved is "only" about 1 Megaton (85 atomics at 12kt each), or triple the size of China's 1966 nuke test. So there would be some Trans-Pacific fallout over the US, but something like 0.5% of what your link estimates (given it's assumption of a 200 megaton regional nuclear exchange).
Which means -- in my purely-amateur opinion -- that the take-away from all this is, the kill-potential of these dirty little atomics the Indos and the Paks have pointed at eachother is dreadful (anywhere from 25 to 100 million dead between the two), but the Trans-Pacific fallout danger to the US is... let's call it moderate to serious, not necessarily grave.
The grave danger to the US comes -- again in my opinion -- not from the danger of a straight-up war between India and Pakistan (even a big one), but from the unpredictable "nightmare" scenarios which could develop: if China moves into the game, or if the Pakis "misplace" some of their atomics into terrorist hands, etc...