Homing sensors issue with parallel port BOB

Update:
the circuit of the first picture actually works, the problem is that when the final black wire is connected to pin 13 the leds behind the sensors dim instead of turning off completely, like if there is some current back feeding from the board to the sensors, but i can see a value change ti the pin value in the HAL Meter..
As a complete "electronic noob" i was thinking about putting a diode to avoid that. Does it make any sense?

Do you have any idea of where i could find a scheme for tat BOB?
I'm trying to understand what is going on by looking at how the various components are connected, but i'm quite lost.
I see that the bottom left pin of the optocouplers (that should be the cathode of the built in LED) is connected via a 1kOhm resistor to the input pins, and i can clearly measure a 2,3kOhm resistance between the top pin (that should be the anode) and GND, and this is exactly the opposite of what i was expecting...

edit:
For example i have a small board with 4 oprocouplers, and those are connected in a more common way.
The 2 input pins go directly into the 2 left pins of the optocpupler (with a 2,2kOhm resistor to protect a small led that shows the logic value of the input), while the 2 right pins are simply connected to ground (the bottom one) and to vcc with a 10kOhm pullup resistor (the top one).

After googling a lot searching for schematics of similar BOBs, PNP NC sensors ecc i came up with this that I THINK should be an approximation of what is going on:

What i understood is that this kind of BOB is just made to have inputs like E-STOPS, mechanic switches ecc, because it simply wants the inputs to be driven by connecting / disconnecting them to /from gnd.

I assumed the position of the led into the sensors: that should explain why the leds only dim: the current from vcc, once the switch gets opened (when the sensor has something near) , can flow to gnd passing through the leds.
Also this would mean that actually those sensors only work thanks to the path to ground that is created through those LEDS...
Does this make sense?
How can i fix this? It looks that i simply have a BOB that is not meant to be used with this kind of sensor..

No, not really.
6 sensors for 2-3 euros each.
I will consider buying some NPN then.
Should i stick with NC? I bought them because i felt they were safer, but in this configuration even if i detach one of them, it doesn't trigger a limit error in linuxcnc.
With NPN however it should be different if i'm correct, the path to ground would be normally closed, and detaching one sensor would definitely change the logic value read by the control.


edit: i'm not fully sure that i really understood how everything works however, if the circuit is the one i drew i still don't understand how the bob is picking the correct logic level of the sensor... the path to ground should always be closed, but i guess that i don't really have enough knowledge of how all the pull-up / pull-down thing works