You could think of dot being a key frame of everything, and a diamond being a keyframe of some things. If in a motion tween you happen to key frame everything, it's still a diamond. But that aside, with classic tweens you can only tween everything between keyframes. With Motion tweens you can tween one thing over some frames, and a different thing over a different range of frames.
It's not a layer thing. If you have a reason to need to tween different attributes over different frame ranges, then use a motion tween. If on every keyframe everything changes, then you may as well use a classic tween, and get the dots.
One thing you will read is that with HTML5 Canvas, classic tweens will perform a lot better. That's because the tween gets converted into one line of JavaScript instead of lots and lots of lines of JavaScript.