Pictures are valuable when researching a subject. Notice the smoke flow in the UP wind tunnel. After hitting the engine's nose, the smoke is not deflected high into the wind. It clings to the locomotive. Why?
From wind tunnel testing of a flat plate (most steam locomotives have a flat plate at the front of the smoke box) we can draw wind streamlines based on how the smoke flows.
Wind tunnel test.

Drawing of streamlines hitting the flat plate.

We can also move a probe from a pressure meter around the object in a wind tunnel.
Here is a plot of pressure around a high-speed train shape.

Reference:
https://www.researchgate.net/figure/Air ... _235609076Right behind the high pressure caused by the engine ramming into the air in front of the locomotive, there is a low pressure (vacuum) that continues along the entire train length.
In the case of the flat plate, the plate shoves the air in front out of the way leaving a low pressure behind the plate. High pressure from the front of the plate and atmospheric pressure around the plate move to fill in the low pressure behind the plate, moving the smoke with it.
It is the same for smoke and steam from a steam engine on the move. The smoke/steam is at atmospheric pressure shortly after leaving the smoke box. The atmosphere, filled with smoke and steam, moves down to fill the vacuum around the locomotive and its train of cars.
A note about vacuum, or low pressure.
A vacuum cleaner does not suck dirt in. It is the atmospheric pressure outside of the vacuum dust collection chamber that rushes in to fill the low pressure inside the vacuum collection chamber that blows the dust into the machine.
It is the same with low pressure around a moving locomotive. The vacuum does not suck the smoke down, higher pressure around the locomotive pushes the smoke into the vacuum surrounding the locomotive (and the train of cars behind the locomotive.)