Possibly, you might be trying to overthink this a tad. Much of the problem is that pilots generally get a fairly superficial introduction to ISA and this ends up causing ongoing confusion.
Reality is that atmospheric conditions change, to a degree, as time goes by. This makes it very difficult to run sensible sums and comparisons for a variety of technical activities.
To allow us to make some useful sense of atmospheric properties (and such are important for a range of activities, including aircraft performance calculations) we need to come up with a proposition which looks at a nominally repeatable (what we might call a "standard") atmosphere. Such a standard atmosphere needs to be similar to some sort of average real atmosphere which has a useful purpose for whatever reason. There are many such "standard" atmosphere definitions (descriptions), used for various purposes, including the one with which we pilots have some familiarity .. which is the ISA (the International Standard Atmosphere). It is important to accept that the ISA is not the be all and end all but has a general usefulness for a number of activities which is why we use it.
The ISA (or, indeed, any standard atmosphere description) presumes that pressure, temperature and so forth vary according to the ISA (or other) definition. You can look up the ISA definition without any problems on the net.
It helps a lot if you think of the ISA definition of pressure height and density height as being the (height) position in the ISA at which the ISA-defined pressure or density for that height occurs. It follows that, for ISA conditions, pressure height will be the same as density height.
Now the real atmosphere will be similar to, but differ to some extent, from the atmosphere according to the ISA definition.
If the pressure in the real atmosphere at some level is so many hPa (we don't use the actual pressure so there is little point in knowing the specific numeric values), and this value of hPa occurs in the ISA at, say, 2500 feet, then what that means is that the pressure height (ie the height at which the observed pressure reading occurs in ISA) on this occasion is 2500 feet. A similar story holds for density height .. For a pilot, that's the extent of things.
So if actual sea level pressure (QNH) is a particular pressure figure other than the pressure value which occurs at zero feet (ie sea level) in the ISA, this just resets the vertical position of the ISA sea level defined position with respect to actual sea level. That's a bit like just running the tape measure up or down a bit and we can run such calculations in the usual manner as you do in your training with Bob.
In the ISA a given pressure height and density height will occur at the same, defined level, as they must if the ISA definition is to hold true. A problem arises, though, if real world conditions are non-ISA .. which is the usual situation.
If, in the real world at a given position and time, the actual temperature differs from what the ISA definition would be at a specific pressure height, then this just varies the density (and density height) from the ISA definition and, to make use of ISA-derived data, we need to run some calculations to figure what the density height is as a level in the ISA defined atmosphere. Again, as pilots, we don't fuss too much about the actual density but run a calculation to figure density height as you do routinely in your theory studies. Again, all we really are doing is moving a tape measure up or down a tad to account for the difference between real world conditions and the ISA definition.
Looking at your specific question ..
If QNH is the actual pressure at MSL, that's the usual story and the temperature at MSL is very high, as it can be at times then why wouldn't QNH be very low? why should it be ? The pressure (and pressure height) relates to the amount of stuff (air) above the point in question. If the temperature varies, that will vary density and the total height of stuff above the point in question but isn't going to cause a problem for the pressure which, as you have indicated, you have just measured or otherwise determined Put another way, why adjust pressure altitude for temperature to obtain density altitude when the QNH used to obtain the pressure altitude surely takes into account the temperature? it doesn't take temperature into account other than the temperature with height variation defined by the ISA definition Put another way again, why doesn't QNH take into account temperature? because that's not how the ISA is defined .. keep in mind that we are talking about an artificially-defined atmosphere, not the real one How is QNH measured? Simplest way, if you can get to sea level, is to have a look at what an altimeter tells you (or measure the actual pressure and run the pressure to height calculations). If you can't get to sea level (which is the usual situation) the QNH can be figured by running some sums based on whatever is the surface pressure you can measure, or infer.
Have a reflect over a cup of coffee and consider how the ISA basically says .. "here at, say, 5000 ft in my artificially-defined atmosphere, the pressure WILL BE so many hPa. If I am at a location (in the real world) where I measure the pressure to be that many hPa, then I WILL DEFINE the ISA pressure height (at that real world location) to be 5000 ft". A similar story applies for density height. The end result is that you need to learn how to run the simplified sums to figure out the correction to apply to real world conditions to figure where the real world actual levels are in the ISA-defined atmosphere.
It can be a tad confusing until you get your head around what the definition is trying to achieve.
