In the most advanced aerosol-climate models it is common to represent the aerosol particle size distribution in terms of several log-normal modes. This approach, motivated by computational efficiency, makes assumptions about the shape of the particle distribution that may not always capture the properties of global aerosol. Here, a global modal aerosol microphysics module (GLOMAP-mode) is evaluated and improved by comparing against a sectional version (GLOMAP-bin) and observations in the same 3-D global offline chemistry transport model. With both schemes, the model captures the main features of the global particle size distribution, with sub-micron aerosol approximately unimodal in continental regions and bi-modal in marine regions. Initial bin-mode comparisons showed that the current values for two size distribution parameter settings in the modal scheme (mode widths and inter-modal separation sizes) resulted in clear biases compared to the sectional scheme. By adjusting these parameters in the modal scheme, much better agreement is achieved against the bin scheme and observations. Annual mean surface-level mass of sulphate, sea-salt, black carbon (BC) and organic carbon (OC) are within 25% in the two schemes in nearly all regions. Surface level concentrations of condensation nuclei (CN), cloud condensation nuclei (CCN), surface area density and condensation sink also compare within 25% in most regions. However, marine CCN concentrations between 30 degrees N and 30 degrees S are systematically 25-60% higher in the modal model, which we attribute to differences in size-resolved particle growth or cloud-processing. Larger differences also exist in regions or seasons dominated by biomass burning and in free-troposphere and high-latitude regions. Indeed, in the free-troposphere, GLOMAP-mode BC is a factor 2-4 higher than GLOMAP-bin, likely due to differences in size-resolved scavenging. Nevertheless, in most parts of the atmosphere, we conclude that bin-mode differences are much less than model-observation differences, although some processes are missing in these runs which may pose a bigger challenge to modal schemes (e.g., boundary layer nucleation and ultra-fine sea-spray). The findings here underline the need for a spectrum of complexity in global models, with size-resolved aerosol properties predicted by modal schemes needing to be continually benchmarked and improved against freely evolving sectional schemes and observations.