Out of curiosity, how many gunshots does she see a year in your part of the woods? I think her odds are a bit too skewed to death given I know a person who survived a point blank shot to the head, a person who was back at work at a near point blank shot to the chest after 6 months, and several who function normally after shots to the leg. Of course my perception could be skewed as the number of gunshot survivors in the USA from both civilian and military situations is likely astronomically higher than Australia. Headshots do tend towards high fatality or vegetable status, but a surprising number of people do live. People survive alot of shootings, but for certain there is a world of difference in surviving a shot to the head with a .22 and a 12 gauge slug. Whether you want to represent this in a game or not is another matter.
EDITED TO ADD: I will also say sometimes people die from almost innocuous things like falling and bumping their head in their houses. So I was not so much posting this to make an argument as much as real life can be a whole lot less predictable than we might want it to be. I can also agree 100 percent with regard to combat effectiveness (which matters more in game maybe than actual survival) and probability of survival with no medical treatment gunshots are going to be combat enders.
I think part of the issue in most games is how the damage ranges are made, as compared to what that amount of damage means. Of course, the granularity of the damage compared to the target's ability to deflect/absorb it matter, but I mean specifically the range of the damage dice compared to expected outcomes. To use a crude example, a crowbar wielded by a 30-year old slender geek is dangerous, but not nearly as scary as one by a 30-year old, 240 pounds of raging muscle. Even if they are both proficient. There's all kinds of scenarios where it doesn't matter (e.g. equally dead or get away), but others where it is the difference between minor bruises and death. The same two people start shooting a .22 at you, proficiency is everything and after that survival chances get down to number of shots and where you get hit (i.e. how high the damage roll is).
This is one thing where I think the skeleton of early D&D is more correct than most later games. Specifically, melee gets that bonus to damage from Strength but missiles get no bonus. If you wanted to make that just a little more deadly and
edge towards realism, I'd suggest a flat damage bonus to melee weapons and increase the range on missiles. Maybe a dagger does 1d4+2 and a short bow goes from 1d6 to 1d8. Even with the flat 1d6 damage option, use 1d6+2 melee and 1d8 or 1d10 for missile.
Don't know what the gun values would be, but suggest that the system concerned about things enough to want to model them should have typical low-end damage values be higher than 1, leaving you with room to drop closer to 1 for the "lighter" weapons, whatever those are. You might even have everyone ignore the first 2 or 3 damage points routinely. A .22 might be modeled by using a range like 1d8 or 1d10 (assuming not using hit locations. Someone might shoot you with a .22 and the hit would be so minor that it wasn't worth recording. Or it might kill you.
Of course, in a war game model, all that gets abstracted out into the chances to hit. So it might be overly fiddly.