I wired up my shiny new KS220M motion sensor switch and played with the motion sensitivity.  I have the motion sensitivity set to the "far range" (25', maximum sensitivity).  The light is set to turn on independent of the ambient light level.

Facing the switch, from the right side of the switch, it seems like it only sees a 90 degree arc standing a few feet away.  From the left side of the switch it seems like it's more like a 150 degree arc.  The two sides definitely don't seem symmetrical.

It's in the garage facing the garage door.  I was able to raise and lower the garage door without triggering the sensor, maybe that's because the garage door is at ambient temperature so the sensor doesn't see any thermal movement.  I was surprised that I was able to drive my car, which was very hot, within 6' of the sensor before the car set off the motion sensor.

Day 2:  I drove up to the light, it didn't turn on, I got out of the car, walked up to the light, I'm 3' from the light, it still didn't turn on, it stayed off until I walked across the path of the light (parallel to the wall that the light is mounted on).

Day 3, KS220M #2:  Mounted in a 4' hallway.  The beam seems very narrow when it worked, maybe 90 degrees.  Because the hallway is 4', you generally walk about 2' away from the sensor as you pass it.  Sometimes it sensed motion and turned the light on, other times it didn't, at one point I was bobbing back and forth in front of the sensor with no luck for a couple of minutes, it finally turned on.

Is there work going on to improve the sensitivity through a software update, or is this the sensitivity we'll have?

On the ambient light level, one suggestion would be to have more meaningful levels, for example:

Instead of 

Cloudy

Overcast (what's the difference between this and Cloudy?)

Dusk/Dawn

Twilight (what's the difference between this and Dusk/Dawn)

Total Darkness

Maybe

Total Darkness

Dusk/Dawn

Indirect Sunlight

Direct Sunlight

It also wasn't clear that the values of these levels are programmable.