2) Inertia is most important at lower frequencies IMHO, higher frequencies don't cause the coil and the cone to move that much because the wave length is way shorter at higher frequencies. Proof of this can be heard when you play through a speaker at low volumes, the highs are there, but to get lows you need to raise the volume.
3) Reactance - That's one important piece of the puzzle...and...I guess you're talking about Inductive Reactance ( since the coil's stray capacitance is so little that it can be surely neglected together with its Xc at audio frequencies ).
Speakers have a nominal impedance, but in the real world impedance changes with frequency, being the square root of the coil resistance squared ( "real" part of the impedance, not frequency dependent ) plus the Inductive Reactance squared ( "imaginary" part of the impedance, frequency dependent ). And, yes, it actually increases with frequency, because Xl = 2Pi*F*L ( Inductive Reactance ).