Pioneer decided to FM modulate 384kbps AC-3 track and mount it on one of the analog tracks. One analog track could have mono soundmix and two digital tracks available for stereo or dolby surround. This way every LD player could play these LDs with audio one way or the other.

I am curious if 384kbps is the limitation of this process? Could a higher bitrate ac-3 track be FM modulated and mounted on one of the analog tracks(or use both tracks for double bitrate)?

Did they simply use 384kbps because the original masters were used?

Dunno about the modulated limits on a digital bit rate, but I do know that it would really have been for naught since I've read that theatrical AC-3 tracks were coded at 320kbps anyway.

The AC-3 RF output actually sends the RF signal from both the left and right FM carriers; the demodulator filters out the 2.3 MHz Left channel signal to leave only the Right channel's 2.8 MHz Quadrature Phase Shift Keyed (QPSK) AC-3 RF signal for demodulation. Pioneer did it that way so they could, if the need arose, supply two seperate Dolby Digital bitstreams or a single high bitrate 640 kbps signal.

What makes you say that? The home version of Dolby Digital doesn't use the compressed theatrical audio. The home formats are encoded from the lossless linear PCM masters, usually with 18 or 20 bit resolution. Dolby Digital is only a delivery format, not a mixing or storage/archiving/mastering format, unlike MLP, TrueHD or DTS-HD Master - and the gold standards, uncompressed linear PCM or Dolby SR encoded magnetic.

The home Dolby Digital system is capable of higher fidelity than the theatrical system, even on LaserDisc, but especially on DVD with the 448kbps rate. And the 640 kbps rate is basically transparent to the source. DTS is the same, with Coherent Acoustics being capable of greater fidelity then the unrelated APT-X100 coder used for the theatrical DTS/Datasat system. The 768 kbps DTS rate sucks though and doesn't even have flat frequency response to 20 kHz - it drops by 3db at 15kHz and has lots of transient distortion and pre-echo.

Per wikipedia, dts in theaters run at 889kbps with 5.0. Dts decoders bass management retrieves the lfe information from mains.

Two dd tracks on ld would require a whole new rf demodulator desing then. None of the rf demds have a switch to chose between 2.3 and 2.8 carriers. But its cheaper to buy a new rf demod than a new player. So it wouldnt be too difficult to adopt to this technology.
They could do dd track and dd isolated sound track or spanish dub(japanese,french,german in overseas) or offer dolby digital enhanced for music lds using combined carriers for 640kbps.

The subwoofer channel in theatrical DTS comes from the surround channels, not the main - the surround signals are rolled of at 80 Hz so the smaller surround speakers theaters typically use won't be overloaded by high levels of bass. The DTS Special Venue format, for IMAX theaters, stores 8 discrete channels on a single CD-Rom but only plays for 55 minutes. The compression system used in the theatrical DTS format is APT-X100 and is in no way related to the home Coherent Acoustics system. APT-X100 is used a lot by radio stations for the studio to transmitter link and automated announcement systems, like at theme parks, etc. The APT corporation has updated it to handle more than 16-bit resolution and higher sampling frequencies, but the upgrades are not compatible with earlier decoders. Terry Beard chose it for the theatrical DTS system because it was just about the only 'good' sounding data reduction system on the market at the time - and it had the lowest distortion. It's bitrate was too high for it to be used in any of the digital television systems then being developed and it's fixed compression rate of 4:1 made it unsuitable for more modern systems like DVD that needed the flexibility of different bitrates depending on the signal or number of channels.

Since Datasat is phasing out the CD-ROM based theatrical system for lossless audio, APT-X100's most well known use will be disappearing.