Re: "...so it could concentrate..."
Engineering explanation coming up, read at your own risk ....
The main control computers used on Juno (and in fact on a lot of space missions) are pretty slow when compared to, for example, your smartphone. They are often based on processor technology that is 15+ years out-of-date; this is basically due to the very long lead-times associated with the development of a deep space mission, coupled with their use of rad-hardened designs that have been proven on previous missions (one thing that deep-space mission designers do not like to do, and that is use the latest gee-whizz technology on their core systems.
So when Juno fires up its main engine, all of sudden the main control systems have to start monitoring a whole bunch of sensors to make sure that everything is going according to plan. For example the system will be monitoring the star trackers to make sure that Juno is not moving off the expected thrust axis - if it does then the AOCS will be commanded to get the spacecraft back on track. Similarly the fuel tanks, fuel lines and the engines themselves will be carefully monitored to make sure that everything stays nominal. This all has to be done very frequently since any thing going wrong will become serious (possibly mission-fatal) really fast!
In order to ensure that the control systems do not overload, it is standard practice in such a situation to shut down anything that is not essential, including all of the science instruments. This ensures that the control system is not having to handle non-critical data streams, and hence minimises the risk associated with the main engine burn.
A further reason for shutting the science instruments down is that they are pretty much useless when the main engine is running - the vibration induced in Juno's superstructure will ruin just about any attempt to gain useful science with anything apart from the magnetometers, and those will be useful due to the engine exhaust planning havoc with the local magnetic field.