Bjork uses an effect that is very simple, yet effective in making stuff sound big (yes, big, not loud!). You might have already tried to achieve similar effects using Stereo Enhancer and Reverb (or either of them alone), but here’s one more thing you might want to try:
Delay left or right
After I wrote the instructions after how I used to do things in FL Studio, eeter pointed out the obivous: instead of doing all of below, just add a Delay DSP (Track DSPs > Native > Delay) to your track, set your feedback to 0%, delay on left or right to 1ms (minimum value), and the other channel to something between 10ms and 40~50ms.
Still, here are some tips for making this work. If your sound has sharp peaks (drums, for example), you usually need a smaller delay (in the area between 10ms and 25ms). If it’s a pad or something equally soft, you can go over 40ms without noticeable echoing. If you overdo it, though, you’ll have an echo, which might be or not be what you want.
Also try muting or unmuting the source in delay DSP for different stereo width and loudness.
Using 2 sends
Another technique to achieve similar results is described below. The main difference is the overall loudness, and also, by muting the source in delay DSP (see instructions further below) on one of the sends, you get more sound “expansion” than with the technique above.
2 sends
You will need a total of 2 sends. I usually use 2 dedicated sends for this. By default you already have S1, so add another one. The two sends will be used to handle one stereo channels per send, so name them “Pull Left” and “Pull Right” (or whatever makes sense to you once you read through this). Pan one send to extreme left, and the other to extreme right.
2 send DSPs
To the track containing sounds that need Bjorkification, add two send DSPs. From the Track DSPs list, expand #Send, and drag #Send to your track’s DPS area twice. The first send DSP from left should not be muted, so enable the “Keep Source” option. From the Receiver pulldown, pick one of the send tracks. The next send should be muted, and the Receiver should be the other send.
It really doesn’t matter which send DPS handles which send, but, they should both send out exact same signals, so you should make sure there are no other DSPs between them. Also, since we send the final, filtered signal, to the send tracks, so you should make sure all track DSPs are before the two send DSPs.
Delay one send track
Final ingredient is delay on one of the send tracks just a little. Switch over to one of the send tracks. It doesn’t matter which, but you should try both and use whatever feels right to you (no pun intended). From Track DPSs list, expand Native, and drag Delay to DPS area. Set feedback for both L and R to 0%, disable Line Sync if you enabled it by habit, and leave everything else at their default values (pan left/right to maximum in correct direction, yada yada). Thanks to Bit_Arts for this tip: Enable Mute Src. (makes a HUGE difference, indeed).
Now, the tricky bit. The L and R delay values should be small, but just enough to significantly enlarge your sound. The smaller the delay, the more subtle the effect. But if you make the delay too long, it will turn into an ugly echo, which you don’t want. I get good results with values from 10ms (very small expansion) to around 40ms(some echo, but still acceptable). Feel free to use longer delays if you actually want the echo (and left-right bouncing effect). I guess it depends on the sound you are trying to process, so experiment with different values until you get the effect you want.
That’s it? You suck!
Hehe, yes, that’s it, and no I do not suck! 
Hey, what’s with this monologue? Anyway, you might think that the effect doesn’t sound as “big” or “woah!” as you thought. There are two possible reasons.
First, you haven’t compared it to normal Stereo Expander output. This really expands the stereo width a lot more without losing any of the intensity. You might simply not have realized it’s greatness (yet). 
Another reason might be that you have killed you stereo sound. This effect doesn’t work as well on stereo sounds as they do on mono sounds. If you have tons of stereo effects like chorus, you might want to consider adding two identical copies on the sends instead of the source track.
Last but not least, check the settings. Make sure that the delay times are not too small or too large.
Update: Hey my sound is too loud now!
I know. Isn’t that great? Anyway, just pull back the volume of the two sends to -3dB, and you should get approximately the original volume.
Comparison
Here is a song using Stereo expander with Surround set to 98%, and stereo setting set to Expand (100%):
http://soundcloud.com/foxbunny/mn01
And here is the same song using the Bjork effect with both sends’ volume adjusted to -3dB:
http://soundcloud.com/foxbunny/mn01-bjorkified
The difference is clearest around 0:50.
Download example .XRNS
(Sorry, no .XRNS until I figure out where to upload it. 
)
.XRNS file for the bjorkified version. The first track “Drone” uses the Bjork effect. It also contains a disabled Stereo expander DSP so you can compare the effects in Renoise.
The example file uses a VST called Synth1. If the the instruments 00 and 03 don’t sound as in the examples on SoundCloud (see above), you can grab the presets.
And I can measure it. It’s really not the same.