Yes the QNH value is based on the atmospheric pressure taken at the measurement site, then adjusted to represent that same atmospheric condition at seal level based on the normal expected pressure change between sea level and the measurement site, and at the standard temperature lapse rate. In other words, if the measurement site was 1000 ft above sea level physically, it's what the atmospheric pressure at the measurement site would be if you were to dig a 1000 ft deep pit at that spot and take the pressure measurement at the bottom of the pit which is physically at true sea level.
The adjustment is based on the lapse rate of the standard atmosphere, so an error is introduced in doing this adjustment if temperatures are significantly different from standard temperatures (59F at SL). Normally not a big deal because everybody in the vicinity is using the same "erroneous altitude", so no harm no foul, but in cold weather this has to be taken into account for IFR operations because the error (altitude reading higher than it really is when temperatures are below standard) becomes significant for the safety of IFR approach operations, and correction tables have to be used, generally when doing approaches in below freezing weather, where the potential high reading error starts to become dangerous in terms of obstacle clearance and decision altitudes.