Wheat hit the nail on the head - pushing the digital audio too hard, low end, conversion problems.
Your first issue (besides a substandard listening environment and consumer speakers) may be the volume levels of the individual audio channels.
First, listen to material other than your project on your audio system, preferably a film with similar characteristics (they both are in the forest and both use synth music); play it as loud as you can (up to 89db) so that it still sounds pleasing without distorting the speakers. This is your mix level. DO NOT CHANGE THIS LEVEL UNTIL YOUR MIX IS COMPLETE.
When you mix you should build around the dialog. If you make the dialog loud, everything else will have to be louder. Start with the dialog peaking at around -18db on the digital meter. As you bring the other audio elements into the soundscape and if the dialog starts getting buried don't make the dialog louder, make the other elements softer.
You also have to be aware of the input and output levels on the audio channel and sub-mix channel plug-ins (such as EQs and limiters) as they can distort, so unless you have the plug-in "open" you won't see it.
The different elements should occupy different sonic spaces. Dialog tends to be centered in the 2kHz to 4kHz area of the frequency spectrum. These freqs can be reduced if the dialog sounds "honky" or increased if it needs some additional intelligability. Again, get the dialog sounding sweet and build around it.
You can use EQ to "carve holes" in the ambient noise; reduce the EQ frequencies occupied by the dialog, you will actually be able to increase the ambient sound volume level a bit without interfering with the dialog.
You can do the same with the music - you can use EQ to "carve holes" in the music to give the dialog more space.
Audio signal (dynamic) compression can be used to increase the "apparent" loudness of audio elements. Just a little compression can get the dialog to pop out of the mix a bit more (too much and you'll increase the noise in the dialog track).
Try to give music and ambiences a wide stereo spread to allow the center to be occupied with dialog and Foley. (Speaking of Foley, you need some footsteps.)
Most mixes tend to be "muddy" meaning that they have too much low end and lower mids. I tend to do a fairly sharp roll-off (18db/octave) with a High Pass EQ starting at about 60Hz.
To sum up briefly, a large part of mixing is SUBTRACTIVE. Begin by taking things away rather than adding; start with a "wall of sound" and pare it away to the essentials, not by eliminating the elements but by reducing their presence.
When outputting the mix you may want to "dither down" the audio. (I don't know if dither is available in programs other than PT, but it makes a difference.) You also want to check your output settings to see if the audio is being converted properly.
There are a thousand more little details that go into a mix - even a simple mix - that are just too complicated to go into here. Last night I put in hour 90 mixing my current feature project and I'm doing a test DVD tonight so I can check the mix on other systems.