This paper presents a multimodal neural recording system incorporating the folded-current-mirror-based reconfigurable signal conveyor for improving hardware availability. The proposed neural recording front-end incorporating the folded-current-mirror-based reconfigurable signal conveyor processes the voltage and current input signals depending on the experiment goal and status of the recording environment. Thus, the number of voltage-recording channels and the number of current-recording channels can be set as desired, maximizing the flexibility of the experiment and the usefulness of the recording system. The fabricated system successfully conducts in vivo experiment in the hippocampus area of a mouse brain. The implemented front-end achieves input-referred noise performances of 2.4 μVrms and 5.19 pArms respectively in the voltage and current recording modes. The whole channel power consumption is 5.718 μW, and only 54 nW is consumed to drive the reconfigurable signal conveyor.