One change I would make could be critical. You have the room, so I would move the small capacitors C1 and C2 as close as possible to the regulator. The very long distances between them and the regulator could have a serious side-effect on the stability of the regulator's output. Consider the full path of current - from regulator to the capacitor and back thru the ground to the regulator. Remember that current flows in a loop. You can't just ignore current when it gets to a ground point. This PC design must have at least 2 inches in the current loops for the capacitors. Other components like the diode D3 are not needed for AC performance and can be placed far away, making even more room for the important bypass caps.
Depending on the dropout rating of the regulator, my calculations show that a 9VAC transformer might be on the edge of the ability to supply 500mA with the rectifier voltage drop (probably about 2 volts total). Not for sure, but it could be an issue for some combinations of regulators, bridge rectifiers, AC power source and transformers. I'd keep the 12 volt transformer if possible, add more heatsinking or airflow. Since you are driving a motor, it is likely that drag or a stalled motor will occasionally demand more current, mostly when starting its motion.
Also, a 2000uF cap should be good for a 12V source, but if going down to a 9VAC, you should have more like 4000uF or more. If your incoming line voltage is low, without an excess of capacitor storage you could starve the regulator and it will produce 120Hz low pulses on the power supply output. I would also recommend at least 25 volts or more for capacitor reliability, 35 or 50 will not hurt anything.
I believe fuses are a good thing. I'd suggest a 2A fuse on the secondary, going into this board.
I'd also suggest a 18...22 volt, 1Watt TVS diode on the secondary to catch any transients that can get thru.