Could the BC327-40-BK and BC337-40-BK be an option?
I have looked at about 6 datasheets, but none of them have noise measurements.
It seems a perfectly nice average small signal device - why do you think it is low noise? Have you seen measurements somewhere?
Now back to circuit layout unless Sharif beats me to it!
Regards
I may have found something.
New device at Mouser Electronics 2SA1312 and 2SC3324.
http://www.toshiba.com/taec/components2/Datasheet_Sync//50/6505.pdf
http://www.semicon.toshiba.co.jp/docs/datasheet/en/Transistor/2SC3324_en_datasheet_071101.pdf
Now that looks more promising. Rbb is claimed 2Ohms! And at high currents (say 5mA) the total noise is comparable to a 15Ohm resistor - about 0.6nV/rtHz.
Where it falls down, and the Hitachi part was so remarkable, is in the 1/f region. At 10Hz, it looks more like 200Ohms worth of noise. Now the ear is not sensitive to quiet low frequency noises, but the the RIAA curve boosts the bass by 20dB, so 1/f comes back into view as a concern.
As you say, the real problem would be finding a matching PNP.