Simulating smoke in Horse and Musket wargames. Waste of time?

 Reading one of the wargaming pioneers always gets one thinking. This time the book was Terence Wise's American Civil War Wargaming (Airfix, 1977). In the rules in this book smoke is simulated and has effects on line of sight and charges. Wise has quite a convoluted system that links smoke movement to a unit's rate of fire (in a way, smoke has a proprietary unit!). What I found really thought provoking is his "Charging through smoke" section that forces a unit to use this fabulous charge deviation template:

Therefore, the unit can get completely lost, charge the wrong unit, expose a flank, etc. How cool!

Is simulating smoke worth the time and hassle? Wouldn't it force the players to establish wind direction, strength and changes like in a naval wargame? However, by not doing it, are we not ignoring an important factor that was often decisive? These are the questions that crowded my head. To see if something (relatively) simple could be done, I drafted this quick smoke rules (the distances are scaled for Simplicity in Practice, but the concept could work in any ruleset).

• Only the densest clouds are tracked. 
• There's no need to track wind direction and speed until someone actually generates some dense smoke.
• When a unit fires in addition to its usual fire dice, it rolls a D10. On a 10 the unit has generated a dense cloud. Place a cotton marker as wide as the unit on it's front. 
• Now is the time to establish wind direction and speed.
• Wind direction: Roll a d8: 1N, 2NE, 3E, 4SE, 5S, 6SW, 7W, 8NW
• Wind speed: Roll 2d6
• At the beginning of each turn all clouds of smoke on the table drift the wind speed in cm in the direction of the wind.
• If firing at a unit though smoke, the unit counts as in cover.
• If moving through smoke, use Wise's deviation template. 
• Smoke lingers until it drifts out of the play area.
• OPTIONAL. A 1 on the "smoke dice" (the d10) dissipates the smoke cloud closest to the firing unit if there is any. 
• OPTIONAL If firing artillery against a town, a 10 on the d10 sets it on fire. The town generates a dense cloud every two turns.
• OPTIONAL. The above assumes a constant wind speed and direction. If you want to go full simulationist, track each turn changes in direction and speed by this method:
•  Wind direction: 1 Backs anticlockwise one step. 2-5 No change. 6 Veers clockwise one step.
• Wind speed: 1 Reduced by 1cm, 2-5 No change, 6 Increased by 1cm

What is your take on this? Is it going to far? Have you ever tried something similar?