@lornop, there's a lot to @Ryuji AB1WX answer, and certainly all valid and worth understanding, but just in case some of it is a bit beyond your current understanding, I'll give you a different take; if you got your Fan Dipole to an SWR of 2:1 or less, THAT'S PRETTY GOOD. In a horizontal or inverted-V dipole antenna, four variables are in play to minimize your SWR and get as close as possible to 50+/-0jΩ and those are:

1. Wire length of each leg of the antenna.
2. The length of each leg (is the feed-point in the middle or offset).
3. The height of the antenna above ground.
4. The angle of separation between the antenna elements.

So in a Fan arrangement you are presumably using antenna wire elements that are approximately 1/4 (or 3/4 for a harmonic band) wavelength for each band you want to be on, and each pair of wires is equal in length (there is a reason to offset the feed-point to raise the "real" part of the impedance, but it doesn't work well as a fan arrangement. The actual length of the legs largely determines the imaginary part (+/-j Ohms)
of the impedance
.

Next is height above ground, and on some of your bands you may find very good real (close to 50 Ohm) impedance, and on lower bands, you may be in the 30-40 Ohm "real" impedance range, and this is due to ground proximity in reference to wavelength of the signal. 

![Impedance vs ground distance][1]

The last factor is the angle between the two radiating legs; as you can see from the chart above, you could have a horizontal dipole where the impedance approaches 100 Ohms, the way to compensate for that is to droop the two the antenna legs toward the ground, symmetrically, and bringing them closer together will lower the impedance. Could you narrow the angle between them on the horizontal plane? Yes, but you will have less of an omni-directional pattern, and your signal will favor one direction over the opposite direction, and if that works for you, then great. You just need to use some antenna modeling software to get a handle on how to arrange the elements. Here is free software with built-in antenna example to get you started: [https://www.eznec.com/]

You might experiment with arranging the element pairs at different angles to each other, keeping each set separated horizontally by 180 degrees, but distributing the pairs around like ribbons on a Maypole if you have the room and the ability to tie them off at various points.

The last point I'll make is in regard to an antenna tuner. While 2:1 SWR doesn't mean your antenna can't radiate most of the energy sent to it, it does mean that your radio transmitter may not be able to send all of its power to the antenna.  An antenna tuner makes a conjugate impedance match between the radio, and the feedline (coax), since while they are both 50 Ohms, if your antenna is not a perfect 50 Ohms, the antenna is reflecting a portion of the energy you're trying to transmit, back to the radio, and an antenna tuner reflects the reflection back to the antenna at the correct phase (that's what a conjugate match is) and increases the power that the antenna must radiate, so even if 10% of the energy is reflected, the reflected reflection is increased roughly by 10%, forcing more power into the antenna. (Total radiated power remains the same, less coax losses) And if there are losses, it is largely due to the resistance and other characteristics of the coax cable, which are typically minimal. But this makes your transmitter operate efficiently because it is getting a 50 Ohm impedance into the antenna tuner, and the tuner is matching the signal energy on the coax that was reflected by the impedance mismatch at the antenna; this re-reflection creates a "larger" signal (greater total energy on the coax because a consistent percentage of it is being reflected by the antenna due to the impedance mismatch) to the antenna, forcing .  So 2:1 and even greater is fine, but in practicality, only if you have a tuner. This can be easily seen if you place a power meter between your antenna tuner and the coax that goes to the antenna; as you "tune" the antenna tuner for the lowest SWR as far as what your radio "sees" into the tuner, you will observe your forward power on the meter on the output of the tuner increase beyond what your radio is putting into the tuner.

  [1]: https://i.sstatic.net/tr93i.jpg