Why, oh why, an ND filter? With tele lens, most of the challenge is having enough light to start with! At the long end, you are starting with F5.8 anyway. Bung on a ND filter and you are most likely below the minimum effective aperture levels needed for AF to work properly. Hence AF difficulties??? Also if MF, what are you using to confirm focus has been achieved - liveview or the optical viewfinder? If the latter, either (1) it may not reveal what's in focus accurately enough (std screens are not really designed for critical focussing) or (2) could be a little off calibration with the sensor. If you use the focus confirm light (which is what I do when MF'ing), it could be picking up a branch rather that the preferred target. If liveview, you would need to use the magnification to see what's going on (not practical though for action shots).
My early long tele shoots were quite so-so. It forced me to considerably upgrade my technique . I use a 55-300 but ideally seek shutter speed 2x 1/focal length if I can get it at desired aperture. If need either ISO400 or 800 for that, so be it. Also use a monopod as much as I can. Try you lens at 300mm with max magnification on liveview and see just how unsteady things can be! My copy did need a little focus adjustment (+30um on my K-x) to get the best out of it. At 300mm, in-focus depth of field is really quite shallow and focus fine tuning does make a difference.
BTW, planes in the air seem a real challenge with haze/contrast, particularly if the camera feels the need to overexpose the scene. And if close and using MF, they rapidly move through the in-focus region. Lot's of misses between the keepers. But can be done. This post on my blog
here is all with 55-300.
Also, even on a tripod, slow shutter speeds and long focal lengths can generate motion blur unless quite careful (heavy tripod, mirror lockup, avoid wind). At 300mm, a liitle vibration is greatly amplified.