I think the point is ( and Vance has a different way of putting it) is that the full vertical component of a downdraft (or the major bulk of it) does not go all the way to the ground, whether it's a weak one or a strong one. Imagine the downdraft as a moving 10 foot wide fluid stream that turns horizontally as it approaches the ground. There is obviously a transitional distance over which that stream turns with the outer edges of the stream turning more gradually and the central part of the stream more abruptly - effectively these outer components will never reach the ground, hence 'do not go all the way to the ground' whereas the central component will possibly go all the way to the ground, but at ground level will have 0 vertical component as it turns more abruptly. There will also effectively be a 'cushion' of horizontally moving air from this transitioning downdraft. All of this is well diagrammed and also shown with radar images in the links I posted. I think the most illustrative images are the radar images showing wind from a downdraft or microburst moving vertically in the higher levels then horizontally at the lower levels. This implies that at the lower levels, closer to the ground, the downdraft no longer accurately describes the direction of air movement. Obviously there are other aspects such as turbulence etc going on as this process occurs, but I think it's pretty clear what Vance means - so yes strictly speaking at least part of the downdraft airmass does reach the ground, but in practice because of the change in vector to a horizontal component no longer acts in a downward fashion on the aircraft. So from the viewpoint of the aircraft and pilot I think Vance is correct that the 'down' part of the draft does not go all the way to the ground, as it becomes a 'sidewaysdraft'.
So I think you are just arguing semantics now. Vance's concept is descriptive from a practical standpoint as the 'down' part of the draft will be much weaker, possibly insignificant to the pilot closer to the ground. One could apply differential equations to this of course but the reality is that a good portion of the downdraft never reaches the ground, and the part that does has 0 velocity downwards as it hits. If I were to only show you the lower level radar images of a body of air that started as a downdraft, you would not be able to differentiate these from radar images of a strong wind blowing across the surface. I think the fact that these images are easily obtained would indicate that this body of horizontally moving air (that used to be a downdraft) extends significantly up from the surface. I will see if I can find info on just how high up this may extend above the surface.